点击显示 收起
1 From the MRC International Nutrition Group, London, and MRC Human Nutrition Research, Cambridge, United Kingdom.
2 Presented at the symposium Fat Intake During Childhood, held in Houston, June 8–9, 1998. 3 Address correspondence to AM Prentice, MRC International Nutrition Group, London School of Hygiene and Tropical Medicine, 49–51 Bedford Square, London WCIB3DP, United Kingdom. E-mail: andrew.prentice{at}lshtm.ac.uk.
ABSTRACT
The fat requirements of children can be judged according to 4 criteria: 1) the possible obligate needs of fat as a metabolic fuel, 2) the provision of a sufficiently energy-dense diet to meet energy needs, 3) the adequate supply of essential fatty acids, and 4) the supply of sufficient fat to allow adequate absorption of fat-soluble vitamins. In these respects the fat requirements of children in developing countries are probably similar to those of children in affluent nations except for the additional needs imposed by environmental stresses, particularly recurrent infections. In many developing countries, the low energy density of weaning foods appears to be a major contributor to growth faltering and ultimate malnutrition. Evidence from doubly labeled water studies suggests that these diets are adequate when children are healthy but fail to support rapid catch-up growth after diarrhea and other infections. The issues in determining and meeting the fat needs of children in developing countries are illustrated with use of detailed comparative dietary data from a rural community in The Gambia and from Cambridge, United Kingdom. The outstanding feature of the Gambian data is the great importance of breast milk as a source of fat and essential fatty acids up until the end of the second year of life. Weaning foods and adult foods contain low amounts of fat, which causes a sharp transition from adequate fat intakes to probable inadequate fat intakes when children are weaned from the breast. The effects of such low fat intakes, particularly in terms of immune function, require investigation.
Key Words: Fat • energy • essential fatty acids • infants • children • breast milk • growth • developing countries
INTRODUCTION
The basic principles of determining the energy and fat requirements of children and the resulting current best estimates are covered in another study in this supplement by Butte (1). This article will therefore focus on those factors specific to developing country environments that may modify a child's energy and fat needs relative to the healthy well-grown Western reference child. It will be assumed that no marked genotype differences affect gross energy and fat needs. This assumption might be wrong [see papers on genetic influences on fat and cholesterol metabolism elsewhere in this supplement (2–5)] and may require modification in the light of further knowledge, but in the absence of any contrary data, such an assumption must be taken as a reasonable working tenet. It will be further assumed that the children of affluent urban dwellers in developing countries are essentially synonymous with the Western child, so this discussion will be confined to poor children at risk of nutritional deprivation.
The published information in this field is largely biased toward the very early years of life when the linkage between nutrient supply, growth, morbidity, and survival is at its most critical. This further concentrates the discussion on early and late infancy, although comments on later childhood are incorporated where possible. Data from our own parallel studies in The Gambia and England will be used to compare and contrast the nutritional conditions in typical developing and developed country settings.
SPECIAL FACTORS MODULATING ENERGY NEEDS IN DEVELOPING COUNTRIES
Environmental stressors—infections
The heavy burden of infectious and parasitic diseases borne by infants and young children in most poor areas of the developing world is probably the major contributor to systematic differences in energy requirements. The potent growth-limiting effects of diseases such as diarrhea and the interactions with diet are well known (6–10). A review by Scrimshaw (11) is particularly useful in summarizing knowledge from a wide variety of sources.
Most communicable diseases affect both the intake and expenditure sides of the energy balance equation to a greater or lesser extent, and compromise growth as a consequence. In many cases the anorexic effects (12), together with reductions in the efficiency of nutrient absorption caused by acute infections (13), persistent gastroenteropathies (14), and gastrointestinal parasites (15), are probably dominant.
Most studies of the effects of infection on energy expenditure concentrated on changes in resting or basal energy expenditure because these are most amenable to measurement. For numerous years there has been a general perception that many infections are associated with a degree of hypermetabolism, which would increase energy needs (16). In fact, this may not be the case for several reasons. First, increases in metabolic rate seem only to occur in response to fever, in which basal metabolism rises by 13% for each 1oC rise in body temperature (17). Further work by Eccles et al (9), who used multiple regression analysis on a detailed longitudinal data set of sleeping metabolic rate measurements in The Gambia, has shown that, when adjusted for fever, an infection such as malaria actually suppresses metabolic rate by 17%. Second, most illnesses are accompanied by lethargy, increased sleep, and periods of bed rest. Theoretical calculations show that these behavioral effects can more than offset any physiologic elevation of metabolism (11). Although we are unaware of specific studies in children, doubly labeled water measurements of adult AIDS patients suffering periods of intercurrent infections confirmed this general principle that, far from increasing energy needs, the net effect of infections can often be to suppress requirements (18) and that weight loss is entirely due to inanition.
Environmental stressors—requirements for catch-up growth
The need for catch-up growth in the anabolic phases following recovery from infection places great temporal demands on energy needs. Detailed theoretical analyses of these needs are available together with quantitative estimates (19). During rapid catch-up growth, the proportion of total energy intake devoted to tissue deposition and remodeling may reach >50% of the total intake (20). As discussed below, the utilization efficiency of this energy may be compromised by a limiting supply of any single nutrient.
The significance of these episodes of weight loss and subsequent catch-up is that they impose periods when energy needs may be exceptionally high and cannot be met from traditional diets. For instance, Ashworth (20) showed that ad libitum energy intakes in children recovering from severe malnutrition increased to 920 kJ•kg-1•d-1 (219 kcal•kg-1•d-1) (ie, 4 x basal metabolic rate) and sustained 15 times the growth rate of normal children for short periods. Such intakes can only be achieved with energy-dense formulations, which in turn impose a need for a high fat content (21). In the discussion below, it will be argued that this episodic need for high-energy foods to support a postinfectious anabolic drive may represent the key area in which traditional diets lead to long-term growth faltering.
Environmental stressors—other nutrient deficiencies
The classic work on protein-deficient pigs by McCance (22) illustrated the profound effect that a single nutrient deficiency can have on the utilization efficiency of dietary energy. This provided a vivid illustration of Kleiber's (23) earlier observation that energy is used less efficiently in the face of specific deficiencies of many vitamins and nutrients. Jackson and Wootton (19) provide further thoughts on the importance of nutrient-nutrient interactions in determining energy needs for maintenance and growth. Generally, the point is that energy will be used with maximal efficiency only if all other necessary nutrients are present in adequate amounts. A rate-limiting deficiency in any one micronutrient or amino acid will decrease the efficiency of energy utilization. Because such deficiencies are common in developing countries, it is reasonable to speculate that energy requirements will be elevated for any given physiologic function, especially growth. Allen (24) summarized many single-nutrient supplementation trials in children in developing countries and pointed out that, generally, the nutrient-supplementation effects on growth are extremely disappointing. She concludes that this is likely due to multiple limiting deficiencies.
Behavioral differences in activity patterns
Behavioral differences in activity patterns, and hence energy needs, vary greatly across different cultural and environmental settings, across different age groups, and between subgroups and individuals. It is therefore only realistic to make a few generic observations.
The general assumption is that children in affluent countries indulge in lower amounts of discretionary physical activity as a result of excess television viewing and computer-game use and of other factors such as predominant travel by motorized transport. Factors such as indoor confinement by inclement weather in northern latitudes or by parental concerns about child safety are frequently cited as being contributory influences in the emerging epidemic of obesity (25). There has been much debate about how early these factors are likely to impinge, with some persons arguing that the effects are noticeable in late infancy, whereas others maintain that the natural ebullience of child play is similar throughout the world and minimizes differences until late childhood. It may be unwise to reach simplistic conclusions in either direction. For instance, the effects of television viewing may be offset by periods of organized sport and recreation in Western societies, and it is notable that duration of habitual television viewing has been found to be uncorrelated with measures of overall activity in children and adolescents (26). It is also possible that chronic energy deficiency in undernourished children suppresses the natural tendency for play.
Comparative data from The Gambia and England on the energy cost of activity in children aged 0–18 mo is presented below. In older children we collaborated in doubly labeled water studies of energy expenditure in Cambridge and Belfast (27) and in 2 villages in rural Senegal with primarily pastoralist and agriculturist livings (B Diahem, unpublished observations, 1994). The estimates of total energy expenditure are compared in Table 1 . An estimate of the energy expended on activity and thermogenesis (ie, EEact, in which the variance can largely be ascribed to differences in activity because thermogenesis is a minor component) can be obtained by subtracting the basal metabolic rate from total energy expenditure (thus, EEact = total energy expenditure - basal metabolic rate). It is clear that in Mbourwaye, the traditional lifestyle of Senegalese children who participate in economic activities, eg, herding animals for the boys and household chores for the girls, is associated with a very high level of physical activity when integrated over the whole day. EEact was up to 3.5 MJ/d higher in the Senegalese children from Mbourwaye than in the British children. However, the caveat concerning heterogeneity of activity patterns is underlined by the apparent large differences in energy expenditure between the 2 Senegalese villages, although the sample sizes are small.
View this table:
TABLE 1. Doubly labeled water estimates of energy expenditure in 9- and 15-y-old British and Senegalese children1
Heart-rate monitoring in Senegalese children showed surprisingly similar results to British children in terms of the proportion of time spent in vigorous physical activity (28). We interpret this as indicating that the high energy expenditure of traditional lifestyles is generated by sustained levels of moderate activity throughout the day.
Torun et al (29) conducted a detailed summary of the available data on energy intakes and expenditure in children from developed and developing countries with some analysis according to nutritional status. In contrast with the clear differences between the traditional and affluent lifestyles shown in Table 1, there appear to be no clear and systematic differences in the literature as a whole. Thus, it is not realistic to make generalized statements about energy needs in developing countries because the heterogeneity between regions and cultures and between rural and urban areas is substantial. The fact that, if anything, energy expenditures are higher in developing countries leads to the counterintuitive conclusion that energy intakes must also be higher. The implications of this paradox vis-à-vis the allocation of energy to metabolic functions, discretionary activities, and growth will be discussed below with specific reference to undernourished Gambian infants.
SPECIAL FACTORS MODULATING FAT NEEDS IN DEVELOPING COUNTRIES
Is there a minimum metabolic need for fat as an oxidative substrate?
Certain organs and tissues are preferential oxidizers of fatty acids as a metabolic fuel (30), a requirement that changes with different physiologic conditions. An obvious example of this is skeletal muscle, in which low-intensity activity tends to recruit type 1 fibers with low glycolytic capacity and high fat oxidation (31). Glycogen depletion after sustained activity also generates an obligate need for fat oxidation (32).
It is theoretically possible that the long periods of moderate-intensity physical activity characteristic of children in traditional settings generate a specific minimal need for fat as a substrate, although the impressive physiologic capacity of altering fuel selection to match fuel supply, as described below, makes this unlikely. Of greater relevance in the current context are the suggestions that many immune cells are obligate (or at least preferential) fat oxidizers (33) and that cytokine actions may suppress glycolytic pathways and hence elevate fatty acid oxidation (34). With respect to the former suggestion, it is noteworthy that lymphocytes taken from specific lymph nodes are maximally stimulated by fatty acids derived from their surrounding adipose tissue (35), suggesting a functional synergy that could imply a direct role for regional fat in modulating immune function. The potential immunomodulatory roles of essential fatty acids (EFAs) are covered below.
The issue of whether acute infections increase whole-body fat requirements is difficult to answer with certainty and there are few data specific to children. Cytokine effects in most acute illnesses lead to marked suppression of appetite. The resultant hepatic glycogen depletion that will occur within 12 h will cause a metabolic switch toward fatty acid oxidation as a secondary consequence of the inanition to minimize gluconeogenic needs. Thus, although cytokine action may well be associated with a reduced respiratory quotient, the effect is most likely to be mediated through inanition and hence would not constitute a specific elevation of fat needs.
In the healthy state, there are again few data on the plasticity of whole-body fuel selection in children, but extrapolation from adult data are almost certainly appropriate. Dietary manipulation studies in whole-body calorimeters show that adjustments in fuel selection can readily accommodate diets ranging in fat-to-carbohydrate energy ratios from 9%:79% to 79%:9% (with 12% of energy from protein) if the body is given 2–4 d to adapt (36). We described elsewhere the mechanism of this adaptation through the oxidative hierarchy (37) and emphasized the central role that carbohydrate plays in controlling the process. If whole-body fat needs could not be reduced as low as the 9% of energy used in these experiments, it would be necessary for some de novo hepatic lipogenesis to occur to supply the obligate fatty acid requirements of other tissues. This would not be detected by indirect calorimetry, which only measures net oxidation rates, and direct measurements of lipogenesis have not been conducted in analogous conditions of high-carbohydrate, low-fat diets (fed at euenergetic concentrations). However, it can be inferred that lipogenesis was not a significant factor because it is a metabolically wasteful process (dissipating 25% of carbohydrate energy), which would increase energy expenditure on high-carbohydrate diets. In fact, whole-body calorimeter experiments show that 24-h energy expenditure is identical across diets with widely ranging fat-to-carbohydrate mixes (36).
Thus, it appears likely that fat balance can be successfully maintained with diets providing as little as 10% of energy from fat. Such very low amounts of fat are claimed to exist under some extreme conditions but are certainly not widely prevalent. Moreover, populations that consumed very low amounts of fat in the past (eg, rural China) are now showing rapid increases in fat intake (38).
Effect of energy density of diets on adequacy of intake
In most foods, energy density is highly correlated to fat content. It is difficult, even with the use of amylase-treated flours, to achieve energy densities much in excess of 6 kJ/g without adding fat to childrens' diets. This is one of the key positive benefits of dietary fat for children in developing countries.
Brown et al (39) reported an excellent study of the effects of increased energy density of foods fed to Peruvian children recovering from malnutrition. They varied the energy density of foods by using nonfat additives from 1.67 to 6.28 kJ/g. Interactions with the number of feeds offered and the time spent feeding were also investigated. The results are summarized in Figure 1, which shows that, although there were some compensatory decreases in the quantity of food consumed, the overall energy intake was more than twice as great with the highest energy density formulation than with the lowest. Other reviews confirm the importance of energy density in achieving high energy intakes in malnourished children (40–42).
FIGURE 1. The influence of the energy density of meals on total energy intake in Peruvian infants recovering from malnutrition. Reproduced with permission from Brown et al (39).
Minimal fat requirements for absorption of fat-soluble vitamins
There is a theoretical possibility that very low fat intakes may impair the absorption of fat-soluble vitamins, which themselves are often in poor supply in traditional diets. In adults, most reviews of fat-soluble vitamin absorption make the implicit assumption that even low amounts of dietary fat will facilitate adequate absorption (43–45). There has been little experimental testing of this in children. A single study from India reported enhanced ß-carotene absorption from green leafy vegetables with fat supplements (46). This will almost certainly depend on which dietary sources of vitamins are being used. We are currently initiating a trial to test the effect of a modest fat supplement given simultaneously with sun-dried mango as the source of vitamin A precursors.
Persistent infection-induced steatorrohea might cause loss of fat-soluble vitamins through a scouring effect, but this would not constitute an alteration in minimal fat requirements because the appropriate solution would be resolution of the infection. In fact, a raised fat intake might even exacerbate the problem.
Essential fatty acid requirements
Uauy et al (47) provided detailed reviews of the EFA requirements of children (48). We will focus only on issues related to developing countries.
Of the many physiologic functions performed by EFAs and their metabolites (eg, prostaglandins, prostacyclins, thromboxanes, and leukotrienes) (47, 48), the immunomodulatory role may be highly significant in the highly pathogen environments of developing countries. The effects of EFAs on immune function were reviewed in detail by Calder (49). Most of the information is from in vitro studies, animal studies, or trials focusing on inflammatory and autoimmune diseases in adults (49), or on lipoprotein metabolism in children (50). It is therefore difficult to make any definitive statements concerning EFAs and immunity in children from developing countries, and it would seem to be an area in urgent need of research.
CROSS-CULTURAL STUDIES IN THE GAMBIA AND ENGLAND
The remainder of this article will use data from our own parallel studies of infants and young children in the contrasting settings of rural Gambia and affluent Cambridge, United Kingdom, to provide a more graphic illustration of the issues involved in considering the fat and energy needs of children in developing countries.
Fat and energy intake of Gambian and British infants
The food and breast-milk intake of Gambian infants was measured as part of a comprehensive research program of maternal and early childhood malnutrition centered in the village of Keneba (51, 52), a rural subsistence farming community in sub-Saharan West Africa with a single rainy season lasting from July to October. The main staples of this community are rice, millet, and groundnuts. Leaves, vegetables, and fruit are available seasonally, fish features regularly in the diet in small quantities, meat and milk are scarce, and palm oil and groundnut oil are used to a small extent. These foods are described in detail elsewhere (53–55).
Infants in rural Gambia are breast-fed to 2 y of age, complementary foods being introduced at the age of 3 mo. The first foods are thin gruels made from only cereal, water (occasionally cow milk is added), salt, and sugar, and are of a low energy and fat content (Table 2). A thicker porridge made from rice and pounded groundnuts is sometimes administered. Cow milk alone is infrequently given to infants <1 y of age; only 57% of infants receive it more than once a week, although it is provided often to children 2 y of age (56). From 6 mo, infants start to share the family food bowl, the most common meals consisting of boiled rice and a sauce made from groundnuts or leaves. Dried fish may be added to sauces in very small quantities, but fresh fish is not given to infants before 9 mo.
View this table:
TABLE 2. Fat content of some common Gambian foods (expressed on a fresh weight basis)1
The most detailed food intake studies in The Gambia were conducted from 1978 to 1980, but the results are still applicable to present times because measurements have shown a broadly similar dietary pattern of infants in the study community compared with measurements recorded 2 decades previously (57). From 0 to 17 mo of age, food intake was weighed for 1 d/mo by Gambian field assistants, as described previously (58). Breast milk was test-weighed for 12 h 1 d/mo (51), and 24-h intakes were obtained using a 2-fold factor previously demonstrated in this community (59). Food and breast milk were analyzed for fat content (51, 53). Fatty acids in breast milk were taken from analyses of Gambian samples (60) and in other dietary items from food-composition tables (61). Food intakes at 2 and 3 y of age were based on measurements conducted previously in the same community (62).
The Gambian results were compared with breast-fed infants in an economically advantaged situation. From 1978 to 1981, parallel studies were conducted in Cambridge, United Kingdom. Weighed records of breast milk and food intake were obtained for 4 d/mo during infancy and follow-up was conducted until 3 y of age, as described previously (63). There are also references made to nutrient intakes obtained in 2 national studies of British infants and young children (64, 65) and to foods given to British infants (66–68).
Total fat intake of Gambian infants over the first 17 mo of age changed relatively little on an absolute basis and therefore declined per kilogram (Figure 2). Fat intake, being provided almost entirely by breast milk, was highest in the first 3 mo. As infancy progressed, increased intake of cereal and groundnut-based foods containing little fat replaced the gradual decline in the consumption of breast milk. The percentage of energy from fat was initially >50%, but declined to 30% by 17 mo of age (Figure 3). Once the infants were fully weaned at 2 y of age, both fat intake and fat as a percentage of energy decreased substantially. The latter was only 15% at 2 y of age, similar to that reported for adult women (60). Food fat was provided chiefly by groundnuts, but cereals were also surprisingly important in their contribution to dietary fat intake because of the relatively large quantities eaten (Figure 4). The few fat-rich foods available for consumption contained oil, but these expensive items were not used frequently.
FIGURE 2. . Fat intake from breast milk and other foods in Gambian and British children. For this and subsequent figures, sample size varies at different ages but is 47 for United Kingdom and >80 for The Gambia.
FIGURE 3.. Percentage of dietary energy from fat in Gambian and British children.
FIGURE 4.. Contributions of different foods to fat intake of Gambian children. GN, groundnut.
Cambridge infants had fat intakes similar to those of Gambian infants over the first 6 mo, but thereafter, fat intakes of Cambridge infants rose despite the marked decline in breast milk intake, which was replaced by other fat-containing foods such as cow milk, biscuits and cakes, eggs, meat, and spreading fats (Figure 2 and Figure 5). From the age of 6 mo, the percentage of energy from fat declined no further than 37% (Figure 3).
FIGURE 5.. Contributions of different foods to fat intake of British children.
Energy intake and growth
The mean estimated energy intakes in The Gambia and in the United Kingdom compared with the FAO/WHO/UNU (69) and the International Dietary Energy Consultancy Group (70) recommended intakes are shown in Figure 6. There is a surprising similarity in energy intakes until 18 mo of age, with the UK intakes exceeding Gambian intakes by only 420 kJ/d (100 kcal/d) from 6–18 mo of age.
FIGURE 6.. Energy intakes of Gambian and British children. Intakes are compared with the FAO/WHO/UNU (69) and the International Dietary Energy Consultancy Group (IDECG) (70) estimates of requirements.
The growth curves for the 2 settings (Figure 7) show an early and profound deviation, however, with Gambian infants falling to 75% of the expected weight-for-age by 6 mo of age. These different growth patterns do not appear to correlate with the estimates of energy intake shown in Figure 6. This is almost certainly attributable to the effects of frequent and chronic infections, particularly diarrhea, which frequently sets up a persistent gastroenteropathy associated with acute phase inflammatory responses (71). These infections limit the efficiency of nutrient absorption and modify the anabolic drive through cytokine actions. As discussed earlier, it may be the failure of low-fat, low-energy-density foods to sustain rapid catch-up growth after temporary weight loss that is the key to the dietary inadequacy.
FIGURE 7.. Growth curves of Gambian and British children.
Studies of energy expenditure
In addition to energy intake studies, we performed detailed investigations of energy expenditure in Cambridge and The Gambia with the use of the doubly labeled water (2H218O) technique (72, 73). Our expectation was that Gambian infants would have lower energy expenditures than their Cambridge counterparts to maintain energy balance with a lower energy intake. One purpose of the research was to investigate the functional consequences of energy deficiency by assessing which components of expenditure (eg, basal metabolism, thermogenesis, activity, or growth) were suppressed and to what extent. The results of this research are summarized in Table 3. Although the levels of energy expenditure were lower in Gambian children on a whole-body basis, we were surprised to find that energy expenditure was virtually identical in the 2 groups when adjusted for the differences in body weight. This implies that in spite of the apparent frugality and low energy density of Gambian weaning foods, the infants can acquire sufficient intake to meet their daily needs with no apparent suppression of basal metabolism or activity.
View this table:
TABLE 3. Estimates of energy expenditure between poor young children in rural Gambia (n = 137) and affluent children in Cambridge, United Kingdom (n = 173)1
This raises the question of why the infants do not allocate to enhanced growth the very small fraction of their total energy budget that would be sufficient to achieve normal growth (1). At least 3 possible answers are possible: First, dietary energy may not be the growth-limiting nutrient. Second, it may be necessary to supply hypernutrition during short anabolic windows after episodes of weight loss to get growth back to its original trajectory. Third, the impaired growth may not be nutritional in origin, but may be mediated through disease processes. Any one of these possible answers challenges the frequently held view that dietary energy is growth limiting in many developing country settings.
Fat quality
Because the fat in the Gambian diet is derived mainly from groundnuts, either directly or through the influence of maternal diet on breast-milk composition (60), the fatty acid profile of the diet is relatively unsaturated. The ratio of polyunsaturated to saturated fatty acids (P:S) rose from 0.46 to 0.66 as infancy progressed (Table 4). Intakes of both total n-6 and n-3 polyunsaturated fatty acids were highest in the early months of infancy, but decreased as weaning commenced; n-6 rose again from 6 mo as groundnuts began to feature in the diet. On the other hand, n-3 polyunsaturated fatty acids showed a further slight decline from 4 mo of age, although the consumption of fish made some contribution after 12 mo of age, despite the small quantities eaten (Figure 8 and Figure 9). These fatty acid profiles changed considerably on the cessation of breast-feeding at 2 y of age, with saturated fats declining to such an extent that the p/s ratio was considerably >1.0 (Table 4). Without breast milk, intake of long-chain polyunsaturated fatty acids (LCPs) also decreased. In contrast, British children of the same age had a much higher intake of saturated fatty acids, and although total n-6 intake was about the same in both communities, British children had a higher intake of n-3 fatty acids.
View this table:
TABLE 4. Estimated daily intakes of types of fatty acids by Gambian and British infants1
FIGURE 8.. Total n-3 fatty acid intake of Gambian children.
FIGURE 9.. Total n-6 fatty acid intake of Gambian children.
Breast-milk fatty acids
Data from many surveys of EFAs in breast milk from mothers consuming traditional diets in developing countries are shown in Table 5. Interpretation of these values, expressed as grams per gram fatty acids, is made easier by reference to Figure 10, which compares total EFA intakes (per kilogram body wt) against the Food Agriculture Organization recommended intakes for Gambian infants. EFA intakes are reasonable in the very early period of infancy, but soon become inadequate, even when infants are breast-fed. The precipitous decrease in EFA intake at weaning reduces the intakes to only a small fraction of the recommended concentration. The very wide range in the n-6 to n-3 ratio of diets, which may have important consequences as discussed by Uauy et al (47), is shown in Table 2.
View this table:
TABLE 5. Major polyunsaturated fatty acids in human milk from mothers consuming high-carbohydrate diets in selected non-western countries
FIGURE 10.. Intakes of preformed polyunsaturated fatty acids in Gambian children. FAO, Food and Agriculture Organization.
Maternal diet exerts one of the most important influences on the fatty acid composition of breast milk (74). Of the fatty acids, those most affected are the medium chain length C18 monounsaturated and polyunsaturated fatty acids (75, 76), with marked geographical variations. For example, the 18:1 in Gambian breast milk reflects groundnuts as the main providers of fat in the diet (60), whereas the 18:2 in Egyptian breast milk is influenced by cottonseed oil in the maternal diet (76) and in South African (77) and Mexican (78) breast milk by maize. In contrast, the lower chain length saturated fatty acids, 10:0 to 14:0, which are synthesized by the breast, are not influenced by dietary intake. The synthesis is, however, increased when the maternal diet is high in carbohydrate and low in fat (74, 76, 77, 79–81), although this may not be so evident when the mothers are in a negative energy balance (60). LCPs are probably less influenced by maternal energy balance or dietary carbohydrate and fat balance because the amounts of 20:4n-6 [arachidonic acid (AA)] and 22:6n-3 [docosahexaenoic acid (DHA)] are generally similar in breast milk from western countries and the developing world (76, 79). The constancy of the ratio of n–6 to n–3 fatty acids in breast milk may reflect protective mechanisms for the infant against alterations in maternal diet (82). A diet high in fish can, however, increase the proportion of DHA in breast milk (80, 82).
Infant needs of LCPs
AA and DHA are the most important LCPs for the developing organs, particularly lipid-rich neuronal tissues, eg, brain and retina, in late fetal and early neonatal life (82–84). Brain growth continues throughout the first year of life and is particularly rapid during the first few months. This growth is associated with increased incorporation of LCPs into phospholipids, primarily in the cerebral cortex (85). LCPs are synthesized from the precursors 18:2 and 18:3, but the capacity for endogenous synthesis is limited in preterm infants and probably also in term infants 8 wk of age (82), or even at 4 mo of age. Although some doubts have been expressed about an absolute requirement for dietary LCPs by term infants (86), there is good evidence of differing LCP status according to LCP intake (84, 87), and dietary 18:2 to 18:3 ratio (84). Breast-fed infants are supplied with preformed LCPs in the milk, but most formula-fed infants do not receive any dietary LCP and their blood and brain concentrations of LCPs are lower than those of breast-fed infants (88, 89). Visual maturation, as measured by visual evoked potential, has also been found to be greater in term breast-fed infants than in those formula-fed infants (90), a finding that correlated with erythrocyte DHA content. Although most of these studies are confined to Western infants, there are no reasons to suppose that these considerations are not equally important in the developing world (87).
There are insufficient data to establish requirements of LCPs for infants, but it is considered appropriate not to depart too far from the profiles in human milk (91). As discussed above, the ratio of n-6 to n-3 fatty acids is important because these 2 families of fatty acids compete for enzymes involved in chain elongation, desaturation, and conversion to biologically active eicosanoids (92). Adequate provision of vitamin E is also required if polyunsaturated fatty acid concentrations are increased.
LCP intakes per kilogram of body weight in The Gambia compared with the Food Agriculture Organization (1994) recommended intakes (92) are shown in Figure 10. It can be seen that a profound inadequacy develops when children are weaned from the breast. The physiologic and health consequences of such an inadequacy in children in the developing world appear to have received little attention.
SUMMARY
The above analysis suggests that minimal fat requirements are not likely to be imposed by constraints related to any obligate need for fat oxidation or by issues related to the absorption of fat-soluble vitamins. However, the need for adequate energy density in foods for young children, particularly for catch-up growth after infections, does have implications for fat content because it is difficult to sufficiently raise energy density without increasing fat intake. The supply of EFAs is also a critical issue; the intakes of precursor and preformed EFAs decrease far below recommended amounts after children are weaned from the breast. Thus, it can be argued that many children in the developing world would benefit from an increased fat intake (perhaps to 20–25% of dietary energy). However, other articles in this supplement highlight the fact that dietary transition has happened extremely rapidly in many communities, with rapid departure from traditional diets. These transitions are associated with rapidly escalating levels of obesity and its comorbidities, which may be especially severe because of the effects of the "thrifty genotype" and "thrifty phenotype." This imposes a difficult challenge in terms of recommending ideal diets.
Much past research on the fat requirements of children in developing countries has focused simply on growth as the only outcome variable. There is a need for a more sophisticated approach, particularly with respect to understanding the physiologic sequelae of marginal and frankly deficient intakes of EFAs.
REFERENCES