Properties of human milk and their relationship with maternal nutrition

Properties of human milk and their relationship with maternal nutrition

Pauline M. Emmett*, Imogen S. Rogers

Unit of Paediatric and Perinatal Epidemiology, Institute of’ Child Health. University qf Brijtoi. 24 Tyndall Avenue. Hrisrol BS8 ITQ. Uh

Abstract

The composition of human milk varies over the course of lactation and in each individual. The volume of breast milk produced is related to the weight of the infant. Human milk is markedly different from cows’ milk, both in terms of macronutrients and micronutrients. This includes the types of fatty acids present and factors affecting their absorption. The types of proteins present and their relative proportions and both qualitative and quantitative differences in the non-protein nitrogen fraction. There is much less lactose in cows’ milk than breast milk and the oligosaccharide fraction is very different. Their are major differences in content and absorption rates of vitamins and minerals from breast milk compared to cows’ milk or formula milk. Vitamin D and vitamin K status are possible problems for the breast-fed infant in certain circumstances. The nutritional status of the mother appears to influence fat concentration and thus the energy content of breast milk as well as its fatty acid composition and immunological properties. There is no coherent evidence, however, that the protein or lactose concentrations are greatly affected. There is some evidence that the concentration of vitamins in the breast milk are influenced by the mother’s intake. Minerals are less variable, with the exception 01 selenium. The response of the infant to human and formula milk differs with respect to endocrine function, stool motility, immune function and renal function. Infant formula milks are designed to mimic human milk as much as possible, but this is unlikely to ever be completely successful. A number of important compositional differences between human milk and formula milk remain. This includes the types and proportions of fatty acids present (which may be of developmental importance), the nature of the non-protein nitrogen component (also possible developmental importance) and the presence of immunoglobulins and fibronectin (which may protect the infant against infection). Q 1997 Elsevier Science Ireland Itd.

Keywords: Breast milk; Formula milk; Cows’ milk; Volume; Fat; Carbohydrate: Protein: Non-protein nitrogen; Micronutrients; Immunity

“Corresponding author.

0.378..3782/97/$17.00 0 1997 Elsevier Science Ireland Ltd. All rights reserved. Pff SO378-3782(97)0005 I-O

S8 P.M. Emmett, IS. Rogers I Early Human Development 49 Suppl (1997) S7-S28

1. Introduction

“There is no better nutrition for healthy infants both at term and during the early months of life than human breast milk. It is a complete food and is species specific.” This is the conclusion of the Working Party of the Panel on Child Nutrition [24]. Indeed, it would be strange if it were not so. Nevertheless, it is quite clear that breast milk is not homogenous. It differs markedly over time and between women.

In order to understand the advantages (and possible disadvantages) of breast milk and breast feeding it is necessary to outline current knowledge regarding the volume and composition of human milk, the way in which these vary within and between mothers as well as the effect of human milk ingestion on various hormones and other intestinal responses in the infant. In this paper we describe the nutritional and immunological properties of breast milk.

2. The volume and composition of human milk

Breast milk volume and composition from the same individual may vary over the course of the day [103], during the course of suckling (difference between fore- and hind-milk) [70] and considerably from day to day [12]. Large” inter-individual variations [l] and changes in composition during the course of lactation [26,61,70,79] complicate the subject. Table 1 shows an average composition of breast milk at various stages compared to that of unmodified cows’ milk and an example of a typical infant formula from standard food composition tables used in the UK [68,111]. The changes in composition are greatest and occur most rapidly during the first week post-partum. The milk produced in the first few days after birth (colostrum) is higher in protein, vitamins A, B12 and K and immunoglobulins than mature breast milk but it is somewhat lower in fat content and hence energy [68,112]. Over the following week the composition of the milk is transitional and can be considered ‘mature’ at about 10 days after birth.

2.1. Volume

The composition and volume of the mature milk varies with each individual mother, the demands made by the infant, the time of the day, the nutritional status of the mother, and whether the milk is suckled or expressed [1,23$X4,91].

There is some evidence that volumes of breast milk produced by mothers in countries where food is scarce are lower than those produced by mothers in Europe and the USA. Milk volumes have generally been assessed by test weighing the infant before and after each feed in a 24 h period and totalling the increments. Conse- quently, the milk intake of the infant is generally recorded as a weight. In the studies quoted below the infants were all exclusively breast fed for the first 4 months receiving no other energy containing food.

Gopalan [43] working with poor mothers in India assessed their mean milk output at 499 g/d (134 g/kg body weight of the infant) at 6 weeks and 527 g/d (109 g/kg)

Table I

P.M. Emmett, I.S. Rogers I Early Human Development 49 Suppl (1997) S7-SB S’J

The average nutrient composition of human milk (in three stages) and unmodified cows’ milk, with onl type of modified infant formula milk for comparison, taken from the standard food analysis tables used 111 the UK

Nutrient Human milk, Human milk, colosnum transitional

Human milk. mature

Cows’ milk, unmodified

Formula milk’;

_--_ Water, g 88.2 Protein, g 2.0 Fat, g 2.6 Carbohydrdte, g 6.6 Energy, kcal 56 Total nitrogen, g 0.31 Saturated fatty acids, g 1.1 Mono-unsatarated fatty acids, g I.1 Poly-unsaturated fatty acids, g 0.3 Cholesterol, mg ?I Total sugars, g 6.6 Na, mg 47 K, mg 70 Cd. mg 28 Mg, mg 3 P. mg 14 Fe. mg 0.07 Cu. mg 0.05 Zn. mg 0.6 Cl, mg N Mn. mg Tr. se. I% N L JLP N Retinal, pg 155 Carotene, pg (135) Vlt D. &g N Vi1 E. mg 1.30 Thiamin, mg Tr. Riboflavin, mg 0.03 Niacin. mg 0.1 TryptI60. mg 0.7 Vit 66, mg Tr. Vit 812. pg 0.1 Folate, kg 2 Pantothenate, mg 0.12 Biotin, pg Tr. Vit C. mg 7

87.4 87.1 87.8 1.5 1.3 3.2 3.7 4.1 .x9 6.9 7.2 4.8

67 69 66 0.23 0.20 0.50 1.5 1.8 2.4 1.5 1.6 1.1 0.5 0.5 0.1

24 I6 14 6.9 7.2 4.8

30 15 55 51 5x I 40 25 34 115 3 3 I1

16 I5 92 0.07 0.07 0.05 0.04 0.04 Tr.

(0.3) 0.3 0.4 86 42 100 Tr. Tr. Tr. (2) I I N 7 IS 85 5X 52

(37) (241 21 N 0.04 0.03 0.48 0.34 0.09 0.01 0.02 0.04 0.03 0.03 0.17 0.1 0.2 0 i 0.5 0.5 0.7

Tr. 0.01 0.06 Tr. Tr. 0.4 3 5 6 0.20 0.25 0.35 0.2 0.7 I .9 6 4 I

a Cow and Gate Premium.

at 4 months. Slightly higher figures were found for under-privileged Bangladeshis [ll], the milk intakes were 151 g/kg at 2 months and 137 g/kg at 4 months. The intakes corrected for the weight of the child were slightly greater than those found in Europe, e.g. 140 g/kg at 2 months and 124 g/kg at 4 months in Copenhagen [70], although the total amount of milk ingested by the Danish children was far greater. This study found that the weight of the infant was the strongest determinant of the

SlO P.M. Emmett, IS. Rogers I Early Human Development 49 Suppl (1997) S7328

variation in milk volume at 2 and 4 months. Neville et al. [81] studied Caucasian women in Denver USA and concluded that “There was a characteristic milk volume for each mother-infant pair that was strongly significantly related to infant weight at 1 month.” A similar conclusion was also reached by Paul et al. [118]. They found a peak intake of 824 g/d in boys and 741 g/d in girls in the third and fourth months and regression analysis showed infant weight to be an important determinant of breast milk volume. Dewey et al. [25] looked at the effect of increasing aerobic exercise levels in the mother on breast milk volume, they found no difference in volume once adjustment was made for the weight of the infant between the exercise and non-exercise groups.

In his studies Gopalan [43] had also looked at the effect of supplementing the mother’s diet during lactation, he found some increase in volume but at the same time there was a decrease in protein concentration and his very small numbers and experimental design make interpretation difficult. Some studies in The Gambia carried out by Prentice et al. [92] looked at the effect on breast milk volume of supplementation of lactating women. A balanced nutritional supplement was provided to 130 women over 12 months, increasing their mean energy intake from 1569 to 2291 kcal/d. There was no effect on breast-milk output compared with retrospective controls at any stage of lactation or in any season of the year.

From the published studies it is difficult to resolve the relationship between milk intake and infant size. Is it the result of lower demand from smaller infants, or has the growth of the smaller infant been limited by a meagre milk supply from the mother? A study on Australian women [93] would suggest that in well-nourished women, at least, the demand of the baby is the driving force, as it appears that the breast is rarely emptied at a feed. Similar studies have not been conducted in the developing world. However, the minimal changes in milk production observed in most supplementation studies might suggest that here also it is the demand of the baby rather than the limitations of the mother that determines milk intake.

2.2. Energy

The energy value of expressed breast milk shows a considerable range mainly due to differences in fat content [23], however the mean energy content used in standard food composition tables in the UK is 69 kcal (289 kJ)/ 100 ml [68]. Obtaining breast milk by expressing it is highly unphysiological and inevitably leads to unrepresenta- tive samples being used in the analysis. A more realistic estimate of the energy content of breast milk as actually ingested may be given by the doubly-labelled water method which can be used to measure the energy expenditure of free living infants. Lucas et al. [66] used this method to study 12 breast-fed infants at 5 and 11 weeks of age. They found the mean energy content of breast milk to be 57 kcal(240 kJ) and 60 kcal (250 kJ)/lOO ml at 5 and 11 weeks respectively, much lower than the level in the food tables. The availability of the doubly-labelled water method has lead to new, much lower, estimates of the energy requirements of infants [90].

P.M. Emmett, IS. Rogers I Early Human Development 49 Suppl (1497) S7-S28 SI:

3. Macronutrients

3. I. Fat

Fat is the main source of energy in human milk and appears to be the most variable of the macronutrients, both within and between individuals and in response to maternal nutrition. The total fat content of breast milk is low at the beginning of each feed, (911 and depends on the extent to which the breast was emptied during the previous feed. As the feed proceeds the fat content rises and can increase as much as four-fold. It also varies according to the time of the day, and with each individual mother and is affected by the type of food she eats [lOO]. The fat in human milk is different from that in the milk of other animals, and as a rule is better absorbed by the infant’s gut. Human milk also contains an inactive form of lipase which once in the duodenum is activated by the presence of bile salts and helps to break down the fats [SO]. The products of this breakdown are particularly well absorbed. L-Camitine, a molecule which is involved with the transport of long-chain fatty acids across mitochondrial membranes, is present in greater concentration in breast milk than cows’ milk. The ESPGAN Committee on Nutrition has recommended that infant formulae should contain at least a similar concentration to that of human milk 131.

Human milk contains both linoleic (Cl 8:2n6) and alpha-linolenic acid (C18:3n3), these fatty acids are essential to an adequate diet. A review by Hemell concluded that infant formula should contain these and that the relative amounts of these fatty acids present is of importance because they compete with each other for the same enzyme during the synthesis of longer chain fatty acids [49]. This was reinforced by the ESPGAN committee report on the composition of lipids in infant formulae which based its recommendation on the ratio found in most human milk samples. lt suggested that a ratio of linoleic to alpha-linolenic of between 5: 1 and 15: 1 should be used in infant formula [3]. Human milk also contains the long chain poly-unsaturated fatty acids (PIJFA) arachidonic acid (C20:4n6) and docosahexaenoic acid (C22:6n3). These fatty acids are not essential in adult diets, but it is likely that pre-term and very young infants cannot synthesise them fast enough from their Cl8 precursors to keep up with the high demand imposed at an early phase of development. They are the only fatty acids utilised by the brain and are important structural components of the membrane systems of all tissues, influencing membrane fluidity and the activity of membrane bound enzymes and receptors. A study which looked at the fatty acid composition of the brain tissues of infants dying unexpectedly compared breast-fed to formula-fed infants and found that those who had been breast fed had a higher mean concentration of docosahexanoic acid than those who had not [30]. Other possible effects on child development are discussed in more detail elsewhere 1421. Most infant formulae do not contain the full range of fatty acids found in human milk. and are particularly low in the long chain PUFAs.

Maternal nutritional status appears to affect the fat and energy concentration of the milk. Michaelsen et al. [70] working in Denmark showed that, in breast milk collected when the child was 4 months old, weight gain in pregnancy was associated

s12 P.M. Emmett, IS. Rogers I Early Human Development 49 Suppl (1997) S7-S28

with fat concentration in the milk, 18% of the variation in fat concentration was explained by this. Fat concentration was higher in the group that gained most weight in pregnancy compared to moderate and low weight gainers. In a study of Bangladeshi women [lo] the fat and energy concentration of the milk was positively correlated with maternal triceps skinfold thickness and arm circumference. Milk production declined as maternal bodyweight fell during the food shortage before the major harvest. The authors concluded that while the lactational capacity of these women remained good (peaking at 750 g of milk per day at 5-7 months post-partum) it was limited by their poor nutritional status. Garg et al. found lower fat concentrations in the colostrum of under-nourished when compared with well- nourished Indian mothers (as judged by BMI) 1371. Again in the Gambia, feeding a high energy supplement to nursing women resulted in a weight gain of 1.8 kg over the course of a year, and an increase in the fat concentration of the breast milk [92]. Karmarkar and Ramakrishnan investigated the relationship between fat in the diet and fat content of the breast milk in 60 Indian women of ‘normal health’ [56]. The milk fat content increased with increasing dietary fat, with an apparent threshold effect at intakes of somewhere between 37 and 60 g of fat/day.

It has been conclusively shown that the pattern of fatty acids in the maternal diet will alter the fatty acid composition of her milk. Milk from vegan women in Britain was found to contain five times as much C18:2 fatty acids as milk from omnivorous women [loo]. Dutch mothers on macrobiotic diets were found to have less saturated fatty acids ClS-20 and more polyunsaturated fatty acids in their breast milk than mothers on omnivorous diets [20]. In a comparison of the composition of breast milk from vegetarian and omnivorous women the concentrations of C16:O and C18:O were found to be negatively correlated with vegetable fat intake, and positively correlated with animal fat and total fat intake [3 I]. The concentration of the essential fatty acid C18:2n6 was positively correlated with vegetable fat intake and negatively correlated with total fat intake. Concentration of C22:6n3 was lower in women who never ate fish.

Stable isotopes have been used in studies by Hachey et al. [46,47] to label palmitic, oleic and linoleic acids. A standard amount of these labelled fatty acids was fed to nursing mothers while on a controlled, chemically defined diet. Breast milk was collected by pump from one breast and the child fed on demand from the other. The results showed that when the mother is in energy balance fatty acids derived directly from the diet account for approximately 30% of the fatty acids found in the milk. Around 60% come from tissue synthesis and adipose stores. On a low-fat calorie- restricted diet a greater proportion of stored body-fat will be used for milk fat synthesis, and in this way the total fat and energy of the diet will indirectly influence the fatty acid composition of the milk. In a carefully controlled study by Insull et al. [51] several different diets were fed to one lactating woman. On feeding a low calorie diet without fat the fatty acid composition of the breast milk reflected that reported for human depot fat. On feeding a high calorie diet without fat the breast milk contained higher concentration of C12:O and C14:O fatty acids than when a maintenance diet (40% of calories from lard) was fed. When fatty acids were synthesised from carbohydrate the proportion of intermediate length fatty acids was

P.M. Emmett, IS. Roger.s / Early Human Developmerzt 39 SuppI (1991) S7-S28 s I ‘i

greater. Harzer et al. [48] fed a variety of iso-energetic diets to 3 women, with different proportions of energy from fat and carbohydrate. Concentrations of C 12:O and C14:O were unaffected by diet, again indicating that these fatty acids art synthesised by the mammary glands.

.7.2. Protein

The average protein content of human milk at the three stages of lactation is given in Table 1. The protein concentration is much higher in colostrum than mature milk but much of it is in the from of secretory immunoglobulin A (IgA) which is probably not absorbed by the gut and therefore is not nutritionally available but is thought to have a function in the protection of the gut from infection [89]. Cows’ milk has a much higher protein content and a very different composition, this is described more fully by Poskitt (861 who has combined various sources to produce some useful summary tables of the fractions of human milk compared to cows’ milk.

Human milk contains some proteins that are synthesised by the mammary gland such as lactoferrin, alpha-lactalbumin and casein, these are called milk-specific proteins. Others are derived from the blood of the mother, such as serum albumin 1651. The concentrations of those secreted in the mammary gland tends to decrease rapidly during the first days of lactation, while those derived from the mother’s blood remain fairly constant.

Milk proteins are of two types; casein, which precipitates out with heat or acid and whey, which remains in solution. Cows’ milk has a much greater casein component than human milk, this leads to a much tougher curd being produced and has resulted occasionally in intestinal obstruction in vulnerable infants 11021.

Approximately two-thirds of the protein in human milk is whey protein. The amino acid composition of human milk whey protein is very different from that of cows‘ milk 1861. The main constituent proteins in human milk whey are alpha-lactalbumin. lactoferrin and secretory-IgA, whereas in cows’ milk lactoglobulins predominate. Lactoferrin is found in greater quantities in mature human milk than in cows’ milk the reason for this is not fully understood Brock [9] has reviewed the literature regarding two possible functions; the regulation of iron absorption and an antimicrobial effect due to the fact that much of it passes through the gut unabsorbed I1 131.

Human milk also contains much more non-protein nitrogen than cows’ milk (possibly up to 25% of the nitrogen). This includes urea, uric acid, creatinine, free amino acids such as taurine and glutamic acid, amino sugars and alcohols, peptide hormones, nucleic acids and nucleotides. The biological significance of these non- protein nitrogen compounds is uncertain, but some are thought to be of developmental importance to the infant [ 13,115]. Non-protein nitrogen is derived mainly from the mother’s blood and fairly constant amounts are present throughout lactation 165). Only a small proportion of the urea in breast milk is biologically available and this depends on its hydrolysis by micro-organisms in the gut [32]. The free amino acid taurine is present in high concentrations in human milk. While this amino acid is not essential in adults, the ability of low-birth-weight infants to synthesise it is known to


be limited. Certain regions of the nervous system contain high concentrations of taurine, and animal experiments have shown taurine deficiency to result in defects in eye development [ 131. Taurine may also be involved in fat absorption, Galeano et al. studied three groups of low birth weight infants fed either with breast milk, or formula with or without added taurine. At three weeks of age the coefficient of fat absorption was higher in the breast-fed and formula with added taurine groups than in the unsupplemented formula group [36]. Evidence suggests that the non-protein nitrogen component may also contain a number of important trophic factors, including epidermal growth factor [13]. Non-protein nitrogen is also found in significant quantities in most infant formulae, its nature and amount being dependent on the method of manufacture and the protein sources used [27].

Michaelsen et al. showed differences in protein concentration in the milk of mothers according to parity [70]. When the infant was 2 months of age 6% of the variation of protein concentration in breast milk samples (n = 74) could be explained by parity and at 4 months 10% could be explained (n = 51). Primipara were higher on average than multipara.

The evidence for any effect of maternal nutritional status on the protein concentration of breast milk is contradictory. It is confused by the variety of methods used to measure the protein component of milk, which may or may not include the non-protein nitrogen component. Garg et al. found significantly higher protein concentrations in the colostrum of well-nourished than undernourished mothers (6.0 and 4.5 g/d1 respectively) [37]. Karmarkar and Ramakrishnan found a significant relation between protein in the diet and milk protein concentration, with (as for fat) an apparent threshold effect at daily protein intakes of around 40 g to 50 g [56]. Other authors have reported low protein concentrations in the milk of undernourished women from several countries including India and Guatemala [80]. However in a study of poorly nourished women from Karachi, Pakistan, the ‘true protein content’ determined by exchange chromatography after acid hydrolysis was similar to those obtained by like techniques in Sweden, Belgium and Japan [54] presumably using milk from well-nourished mothers. In an early study of low income Yoruba women, in Ibadan, Jelliffe found the protein content of the milk to be similar to values obtained for well-nourished western women, with no apparent fall on prolonged lactation [53]. In addition, Brown et al. found no relationship between anthropometric measurements of Bangladeshi women and the nitrogen concentration of the milk [ 1 I].

The results of protein supplementation trials are no more consistent, and their interpretation is complicated by the difficulty of ensuring that the supplements augment the protein in the maternal diet, rather than replacing it. Prentice et al. found a 7% increase in the protein concentration of the milk on supplementing the diet of lactating Gambian mothers [92]. Motil et al. found milk total nitrogen and protein nitrogen concentration to be negatively associated with nitrogen balance [78]. In a study of three well-nourished Swedish women increasing the protein intake from 8 to 20% of energy increased the total protein and non-protein nitrogen concentration of mature human milk and the 24 h milk protein output [33]. However, in a protein supplementation study carried out by Gopalan, an increased output of milk was

P.M. Emmett, I.S. Rogers I Early Human Development 49 Suppl (1997) S7-S28

associated with a fall in protein concentration, with the result that the 24 h protein output was not significantly altered [43,44].

3.3. Carbohydrates

Lactose is the principal carbohydrate in human milk (mean of 46 samples 5.6 g/100 ml on day 4 of lactation and 6.9 g/100 ml on day 120) [18]. It is not an essential nutrient to the human diet but may have some special functions in breast milk including aiding calcium absorption, developing the acidity of the gut and determining the nature of the gut bacterial flora [l&l 161. There is a greater concentration of lactose in human milk than cows’ milk and although infant formulae have been modified to have a similar level of total sugar to that in breast milk other glucose polymers rather than lactose are often used to achieve this.

The lactose concentration of human milk appears to be fairly insensitive to changes in diet and nutritional status, although Prentice’s Gambian study [92] found concentrations which were slightly higher than the Western norm prior to supplementation. After supplementation the lactose concentration had decreased and fat concentration increased resulting in the total energy content remaining the same. Possibly this reflects the fact that lactose is metabolically more economical for the breast to produce.

Oligosaccharides are present in a reasonable amount in colostntm and to a slightly lesser extent in mature milk (mean 2.1 g/100 ml on day 4 and 1.3 g/100 ml on day 120) [ 181. A further 1.2% of the total carbohydrates in colostrum are monosac- charides mainly glucose and fucose, this declines in mature milk. Oligosaceharides along with lactose facilitate the growth of the bifidus flora in the gut which helps to protect the breast-fed child from gastrointestinal infection [IS]. They also have a role in inhibiting bacterial adhesion to epithelial surfaces thus preventing infection [ 171.

4. Micronutrients

A plentiful supply of breast milk from a mother eating an adequate diet should provide all the infant’s requirements of vitamins, minerals and trace elements. in general they have a high level of bio-availability so that even at a low concentration they can be well utilised [64]. A typical micronutrient content, the mean of several samples of expressed human colostrum, transitional milk and mature milk, is shown in Table 1, these values are from standard food tables and would be the ones likely to be used in assessing the dietary intakes of infants [68]. However, the micronutrient content of breast milk may vary between individuals, with maternal dietary intake or nutritional status, The effect is very variable depending on which micronutrient is under consideration. In general the concentrations of water soluble vitamins appear more responsive to maternal dietary intake than the concentrations of minerals and fat soluble vitamins. For many vitamins it seems that the concentration in human milk will rise in response to increasing maternal intake until a plateau is reached, but even


when considering a single micronutrient the evidence of the effect of maternal nutrition is often contradictory.

4. I. Water-soluble vitamins

Some water soluble vitamins such as thiamin, niacin, vitamin B6, folate, pantothenate and biotin increase in concentration as lactation proceeds from the production of colostmm to that of mature milk. Others such as vitamin B 12 and vitamin C decrease in concentration, while the riboflavin concentration remains the same. Some are affected by the diet of the mother.

Dietary supplementation of lactating Gambian women had a considerable effect on the breast milk content of water-soluble vitamins [92]. Supplementation significantly increased the concentrations of thiamin, riboflavin, nicotinic acid and ascorbic acid. In an American study vitamin B6 intake of the mother was a strong indicator of infant vitamin B6 status [55]. Similarly, a survey of South Indian women found a significant positive correlation between the pantothenic acid, riboflavin, nicotinic acid, ascorbic acid and thiamin content of the diet and the breast milk concentration of these vitamins [22]. In much of the third world the vitamin C intake varies seasonally with the availability of fresh fruit and vegetables. The vitamin C content of the milk of poorly nourished women in Botswana was found to be higher in the wet season (2.7 mg/ 100 ml) than in the dry season (1.7 mg/ 100 ml) [ 1071. In some societies dietary taboos during the puerperium might have a marked effect on the water-soluble vitamin content of the breast milk. In a study in Malaysia many mothers were reported to exclude most fruit and vegetables from the diet for 30 to 40 days after giving birth, resulting in a major drop in their vitamin C intake [71,72]. Similarly, a study of the food behaviour of Chinese immigrants in London found that many women felt it necessary to avoid fruit and vegetables in the first two weeks after delivery [117] this is due to cultural beliefs which classify foods according to a system of ‘hot-cold’. Fruit and vegetables are considered to be cold and a possible cause of diarrhoea.

Numerous studies have found the vitamin B12 content of breast milk to be associated with maternal diet. Specker et al. found the vitamin B12 content of the milk of strict vegetarian women was lower than that of omnivorous women, and that it was inversely related to the length of time the mother had been on a vegetarian diet [105]. The findings of Dagnelie et al. were similar [20]. They compared the vitamin B12 content of the milk of women on macrobiotic diets (containing little meat, fish or dairy products) and on omnivorous diets. The B12 content of the milk was lower in the women on macrobiotic diets, and was positively correlated with meat and fish consumption. These results led the authors to counsel against complete abstinence from meat and fish consumption during pregnancy, so as to avoid the risk of nutritional deficiency. However, a longitudinal study of marginally nourished Brazilian women over the first 9 months of lactation found no relationship between maternal blood and serum concentration of vitamin B12 and folate and the vitamin B12 and folate content of the breast milk [26]. Furthermore, folate supplementation during pregnancy had no effect on breast milk folate content at any stage of lactation.

P.M. Emmett, IS. Rogers I Early Human Development 49 Suppl (1997,) S7-S2X SI

Several studies of apparently well-nourished women in industrialised countries have also found no correlation between maternal serum and milk folate concentration [99,104,1 lo]. There is evidence that folate is preferentially partitioned to mammaq tissue, maintaining the folate content of the milk at the expense of the mother’s folate status 1691.

4.2. Fat-soluble vitamins

The concentration of vitamin A in human milk decreases over the course o1- lactation [ 16,191. Several reports have indicated an association between the vitamin A content of the diet in pregnancy and lactation and vitamin A concentration in the milk. The vitamin A content of the milk of poorer populations in developing countries such as India, Ceylon and Jordan, where intakes are marginal, is lower than in North America and Europe [54]. The vitamin A content of the milk of normal Swedish mothers was found to be higher than in non-privileged Ethiopian mothers 1391. However, Garg et al. reported no significant difference in the vitamin A content of the colostrum of well- and undernourished Indian mothers [37].

The results of supplementation trials are inconclusive. A rise in the breast milk content of vitamin A in Central America was reported after the introduction OF fortified sugar [6]. Kon and Mawson found vitamin A supplementation before and after parturition and later during lactation to result in the secretion of milk richer in vitamin A than normally produced 1581. Stoltzfus et al. ]109] working with 153 Indonesian mother infant pairs carried out a randomised, double-blind trial of a high dose vitamin A supplement or a placebo given to the mother. The mothers’ mean serum retinol concentration was higher in the vitamin A group at 6 months as was the mean retinol concentration in the breast milk of this group. The infants of the supplemented group were less likely to have a low serum retinol level (15% ) compared to the placebo group (36%, p < 0.005) and had significantly higher vitamin A stores. However, several studies have found no effect of vitamin A supplementation on breast milk composition [8,114]. Similarly, Chappell et al. found no association between the reported maternal intake of vitamin A and carotene and the composition of the breast milk in well-nourished Canadian women [lS].

There have been reports of clinical rickets in breast-fed infants often associated with fad diets [28] and in countries where climate or custom lead to low levels of exposure of the child or the mother to sunlight infant serum concentrations of 25-hydroxyvitamin D (25OHD) may be sub-optimal [4,106]. A randomised double- blind study was carried out by Rothberg et al. [95] on term mother-infant pairs in the winter months in South Africa. The mothers, who were all well-nourished, were randomised to a placebo or one of two vitamin D supplemented groups, their infants were not supplemented. After six weeks mothers in the supplemented groups had higher serum concentrations of 25-OHD than mothers receiving placebo, the infants in the supplemented groups also had higher serum concentrations of 25-OHD than those in the placebo group. However a control group of infants who had received direct supplementation with vitamin D had even higher serum concentrations. Ala- Houhala 141, working in Finland in the winter when sunlight levels are low, also


looked at the effect of supplementation with vitamin D on the serum 25OHD status of mothers and their breast-fed infants. In this study supplementing the mother with 10 IU/d vitamin D had very little effect on the infants 25OHD levels, however supplementing the infant with either 400 or 1000 IU/d had considerable effect. During warm weather the infant’s skin exposure to sunlight ensured adequate levels of 25OHD but in winter many of the unsupplemented infants had levels below that considered to put them at risk of developing rickets. With the current trend to protecting children from the risk of skin cancer by reducing their exposure to sunlight there may be an increase in the incidence clinical rickets in breast-fed infants.

Vitamin K is a possible problem for the breast-fed infant. Human milk contains only a low concentration of vitamin K and there is strong evidence of increased incidence of late haemorrhagic disease in breast-fed infants [62]. Single pharmaco- logical doses of vitamin K to the mother have been shown to cause a short term rise in the vitamin K content of the milk, although the concentration returns to normal within a few days [45]. However, Pietschnig et al. found long-term vitamin K supplementation at a level nearer the recommended daily amount (RDA) to have no effect on the vitamin K content of breast milk, and also found no dose-response effect of maternal dietary intake [85]. This raises the unresolved question as to why nature has ensured that the young infant is kept in a vitamin K deficient state.

Vitamin E concentration is higher in colostrum than mature human milk, and there is little evidence of an effect of mother’s diet on the concentration. Garg et al. found no difference in the vitamin E content of the colostrum of well- and under-nourished mothers [37], and Chappell et al. found no correlation between vitamin E concentration in milk and maternal intake [15]. However, a single case report has documented an effect of high maternal intakes of vitamin E (approximately 27 mg/day), resulting in a high vitamin E content of the breast milk [5].

4.3. Minerals

The bioavailability of most minerals in breast milk is much higher than from cows’ milk or infant formula. As with vitamins the effect of maternal diet on the content of minerals seems to depend on which mineral is being considered.

The concentrations of most minerals in the breast milk remain fairly constant throughout the course of lactation. The exceptions are zinc, copper and iron. These minerals have their highest concentrations immediately after parturition, and fall for several months thereafter [26,61,79]. A comprehensive review of the neonatal metabolism of copper and zinc was undertaken by Aggett [2] he suggests that low birthweight infants are more vulnerable to deficiencies of zinc and copper than those born in the normal birtbweight range. Maternal intake of these minerals does not appear to affect the breast milk concentration. No relationship was found by Donangelo et al. between the iron and zinc content of the breast milk, and blood and serum indicators of maternal nutritional status for these minerals [26]. There was no significant difference in the iron, copper and zinc content of the colostrum of well- and undernourished Indian mothers [37]. In another Indian study there was no difference in the iron content of the breast milk between groups of women with iron

P.M. Emmett, IS. Rogers 1 Early Human Development 49 Suppl (1997) S7-S28 s l(J

intakes ranging from 8.0 mg to 40 mg per day [56]. Mozer and Reynolds found no relation between maternal zinc intakes and zinc concentration in human milk (741 and also found no relationship between plasma zinc and milk zinc concentration jn a comparison of American and Nepalese women [76].

Most zinc supplementation studies have likewise failed to convincingly demon- strate any effect on breast milk zinc concentrations. Moser-Veillon and Reynolds [77] studied 40 American women from 0 to 9 months lactation half of whom received a daily supplement of 25 mg zinc. They observed no difference in milk zinc concentrations between the supplemented and placebo groups. Similarly, Moore et al. found no changes in milk or serum zinc concentrations after one week of daily supplementation with 50, 100 or 150 mg zinc 1731. Krebs and Hambridge found that giving a daily supplement of 15 mg zinc to lactating women signifuxntly reduced the fall in milk zinc concentrations from months 1 to 9 of lactation compared to an unsupplemented group, although the plasma zinc values of the 2 groups were similar during this time [59]. The women they studied failed to meet the American RDAs for zinc from their diet, and the authors suggested that the more rapid fall in milk zinc concentrations seen in the milk of the unsupplemented mothers reflected a zinc- deficient diet. The differences in milk zinc concentrations between the supplemented and unsupplemented groups were relatively small up to 6 months, but averaged around 50% from 7 to 9 months post-partum. This led the authors to suggest that they might become nutritionally important where breast milk makes a large contribution to the diet for more than 6 months. However, from 6 months the number of women remaining in their study was greatly reduced.

In a recent study in Finland lactating women were supplemented with either 0, 20 or 40 mg of zinc daily [98]. The 20 mg supplement had no effect on milk zinc concentration, but the 40 mg supplement reduced the fall in breast milk zinc concentration over 7.5 months (p = 0.02) although there were only 5 women remaining in the high supplement group by this time. In an earlier study 33 American and 30 Egyptian women were given either a placebo or a 50 mg zinc supplement daily for 34 days at the 7th to 9th month of lactation 1571. The decrease in milk zinc concentration over the study period in the American group was significantly smaller in the supplemented women. However, there was no difference between the 2 groups of Egyptian women. Furthermore, the authors noted that the lesser fall in milk zinc concentrations in the supplemented American women may have been due to the fact that milk zinc concentration were initially lower in this group.

The close control of the concentrations of calcium, phosphorus and magnesium in maternal serum would make it seem unlikely that maternal nutrition would significantly affect their concentration in human milk. Garg et al. found no effect of maternal nutritional status on the milk concentration of calcium and magnesium [37]. While the milk of Pakistani women tended to have a lower mineral content than that of women in the UK (phosphorus, sodium, potassium and magnesium) the calcium content was the same [79]. Moser et al. found the milk of Nepalese mothers to have a similar calcium content to that of American mothers despite a considerably lower dietary intake (482 vs. 1144 mg/day) [75]. However Dagnelie et al. found the milk from mothers on macrobiotic diets to contain less calcium and magnesium than milk

s20 P.M. Emmetr, IS. Rogers I Early Human Development 49 Suppl (1997) S7-SZ8

from mothers on omnivorous diets [20]. Magnesium, but not calcium concentrations were positively related to meat and fish consumption. In addition, studies of mothers in the Gambia and Zaire, where calcium intakes are relatively low, found the breast milk calcium concentrations to be E-20% lower than those of mothers in Cambridge, where calcium intakes are high [63,88]. However, it is not clear how far the calcium content of the breast milk was affected by the diet, and in Zaire no influence of maternal nutritional status on breast milk mineral concentration was observed.

On the whole the studies have shown that the maternal diet has little effect on the concentration of calcium and magnesium in the breast milk. However, there is good evidence that the selenium content of breast milk is affected by maternal diet. Moser et al. found dietary selenium, plasma selenium and milk selenium all to be lower in Nepalese than American mothers [76]. Funk et al. found a seasonal variation in the selenium content of the breast milk of Gambian women, with lower milk selenium coinciding with food scarcity in the rainy season [35]. The selenium content of the soil and hence the food supply has large geographical variations, as does the selenium content of mature human milk [80]. A study of Finnish mothers would suggest that the chromium content of the breast milk is unaffected by the maternal dietary intake [W.

5. The infant’s response

5.1. Endocrine responses

In order to assess the influence of non-breast-milk feeds on the normal term infants’ endocrine responses, a comparison was made by Lucas et al. between 43 who were breast-fed and 34 who were bottle-fed at 6 days old. All babies were fed at four hourly intervals. The study showed significant changes in plasma concentrations of insulin, motilin, enteroglucagon, neurotensin and pancreatic polypeptide after feeding in the formula-fed group which were reduced or absent in the breast-fed group [67]. The authors suggested that the differences in insulin response may explain differences in subcutaneous fat deposition between breast- and formula-fed infants.

5.2. Effect on the gut

Studies have shown that the time taken by the new-born infant to pass the first yellow stool is far faster in bottle-fed than in breast-fed babies. A prospective study of 150 babies born at term and weighing 2500 g or more showed that 65% of the bottle-fed babies had passed their first yellow stool by the third day of age compared with only 32% of the breast-fed (p < 0.0001) [97]. This study also showed that the babies fed by bottle excreted more bilirubin and consequently had less jaundice than the breast-fed infants.

Most of the research concerning the effect of breast milk on the gut has involved the microflora in the gut, which show lower bacterial counts in the stools of

P.M. Emmett, IS. Rogers I Early Human Developmer~t 49 Suppl (1997) S7-32X S?I

breast-fed infants, and lower incidence of diarrhoea or gastro-enteritis from most, but not all, pathogens [82].

Differences in intestinal flora between breast- and bottle-fed infants are reflected in different faecal concentrations of short chain fatty acids [29]; these may be important for the health of the colonic mucosa. There are differences too in faecal flora according to the type of formula milk used. Merely changing the iron content of infant formulae results in differences in the colonisation and growth of various bacteria [7].

The differences between the response to cows’ milk formula and human milk is not surprising. As Jackson and Golden [.52] pointed out, “the cow is a ruminant. and cows’ milk has evolved to promote bacterial growth in the upper small bowel, whereas human milk has evolved to discourage bacterial growth”. They go on to point out that the constituents of the two milks show that their differences can be accounted for in terms of this difference in function. For example, in human milk there is present specific IgA, maternal lymphocytes, lactoferrin and lysozyme. In comparison, in cows’ milk specific IgA and maternal lymphocytes are absent. lactofenin and lysozyme are present only at trace levels, and whereas human milk has poor buffering capacity, cows’ milk has good buffering capacity.

5.3. Immunio

There is considerable evidence to suggest that infants who are breast fed are less likely to develop infections [40,41]. This is partly because the feeding of foods other than breast milk increases the risk of exposure to infectious organisms, but also because breast milk seems to have an effect on mucosal immunity. A number of studies have been carried out.

One study in Cambridge [87] compared ten breast-fed and twelve formula-fed infants at 6 and 12 weeks of age. The amount of IgA excreted in the urine was three times higher among the breast-fed babies than the bottle-fed. This was not related to the amount of IgA in the breast milk but breast feeding seems to increase the local production of the secretary IgA into the urinary tract during early childhood, thus resulting in potential protection from infection.

A comparison of the IgM and IgA concentration in the saliva and nasal secretions of fifteen breast-fed and fifteen bottle-fed British babies, showed that total IgM and IgA concentrations and IgA antibodies to E. coli were higher in the saliva and nasal secretions of breast-fed infants at 6 days, but thereafter there was a rapid increase in concentration of IgM and IgA in both breast- and bottle-fed groups with little difference between the two. There were no differences between breast- and bottle-fed babies for total IgG, or specific IgG, IgM or IgA antibodies to tetanus toxin, or IgG and IgM antibodies to E. coli. The authors suggest that breast feeding enhances the secretory immunity in the early neonatal period only, but that by six weeks local antigens are the main source of stimulation for the production of immunoglobulins in the respiratory mucosa [ 1081.

A study was undertaken in New York into plasma fibronectin concentrations. comparing 26 breast-fed infants with 27 formula-fed infants, Fibronectin acts as a


non-specific opsonin by attaching itself to invading bacteria making them more attractive to phagocytes, thus modulating reticuloendothelial clearance of bacteria and intravascular debris. Consequently it has a potentially important role in host defence against infection. Human milk contains fibronectin. The authors showed that at 2 to 5 days of age there were significantly higher plasma concentrations of fibronectin in the breast-fed infants (237 mg/l) compared with the formula-fed infants (171 mg/l) [34].

Maternal IgG transferred across the placenta to the fetus during intrauterine life represents an important component of the neonatal immunological defence mecha- nism against infection. Nevertheless further benefits can accrue after birth from breast feeding. An Italian study looked specifically at 11 healthy term infants, and at the breast milk from the mothers. It showed that although the amount of total IgG in the milk was extremely low there were significant amounts of specific antibodies [38].

A study by Pabst and Spady [83] suggested that babies who had been breast fed responded to vaccinations with higher antibody levels. A Canadian study [loll, however, in trying to replicate this study, tested 408 infants at 7 months of age one month after they had completed 3 doses of Huemophilus injluenzae B vaccine. The authors found no relationship whatsoever between antibody response and duration of breast feeding.

Maternal nutritional status has been shown to affect the immunological properties of breast milk. Saha et al. studied the composition of the colostrum in well-nourished (BMI > 20 kg/m2) and undernourished (BMI < 20 kg/m’) Indian mothers [96]. The total cell count was higher in the colostrum of the well nourished mothers as was the level of secretory IgA [37]. In this study no significant difference was found in the lysozyme content between these two groups of mothers. However, this may have resulted from the definition of nutritional status purely in terms of BMI. Sue-Joan Chang used a more stringent definition of nutritional status, which also included serum albumin levels. She analysed sequential samples of milk from 60 well- and malnourished Chinese women. During the first 7 days of lactation and at most stages up to 8 weeks thereafter the mean concentrations of IgA, C3, C4 and lysozyme in the milk from the malnourished group were approximately half that found in the well-nourished group [14]. She concluded “The findings of this study are highly suggestive of a detrimental effect of maternal malnutrition on some protective substances in breast milk”.

5.4. Renal solute load

Davies showed that in very young infants, up to three months of age, plasma osmolality was related to feeding practice [21]. Plasma osmolality reflects the capacity of the kidney to handle the dietary solute load. Breast milk provides a lower dietary solute load than cows’ milk, normal infant formula or solid foods. The immature kidney has difficulty maintaining plasma tonicity when faced with a high solute load. In this study the breast-fed group had a lower mean plasma osmolality than the formula-fed group and those fed both formula and solids had an even higher mean osmolality. It was pointed out that even a small loss of fluid due to infection or anorexia could lead to dangerously high blood osmolalities in the young infants. This

P.M. Emmett, IS. Rogers I Early Human Development 49 Suppl (1997) S7-S28 S?’

has led to the recommendation that the introduction of unmodified cows’ milk and solids to infant diets be delayed [24].

6. Discussion

The differences in composition between cows’ milk and human milk are related to their evolution for the feeding of very different animal types 1.521. The human infant has a very slow growth rate compared to animals other than primates. This influences the concentration of nutrients the milk contains. Other properties are related to the type of gut the human has, the types of diseases which affect humans and the basic immunological mechanisms which are species specific.

Cows’ milk has a lower concentration of lactose, polyunsaturated fatty acids and free amino acids as well as vitamins A, D and E than human milk. However it has ii higher concentration of potassium, sodium, calcium, phosphorus, zinc, manganese. and folic acid. Other components, such as iactoferrin and transferrin, although present in both are in a different ratio to each other thus altering their effectiveness. The rate of absorption of nutrients such as calcium, iron and zinc can be affected by these differences. Formula manufacturers are continuing to research into the modifications necessary to make cows’ milk similar to human milk, however there are so many subtle differences, some of them relating to a specific mother-child pair, that it is difficult to envisage that this can ever be completely successful. For instance, some formulae now have taurine added and others have very long-chain PUFAs, however manufacturers have found that problems with keeping quality can then arise. Other factors present in breast milk such as nucleotides, polyamides and glycosamine are little understood and may well be found to be important in the future.

There are major differences in the consequences of breast feeding compared to bottle feeding as measured at the cellular and physiological level. The key question is whether they result in differences in health, growth and developmental outcome ot the infant, and our reviews of the literature show that breast-fed babies are indeed protected from some infections [40,41] and appear to have enhanced cognitive development compared to bottle-fed babies [42].

One must not lose sight of the fact, however, that where the nutrients supplied by the breast milk are not derived from the diet, they must be provided from maternai reserves. The possible long-term consequences for maternal health of nutrient depletion by lactation have been under-researched but should always be remembered in formulating policy [94].

References

[ 11 Adock EW, Brewer ED, Caprioli RM, West MS. Macronutrients. electrolytes and minerals in human milk: differences over time and between population groups. In: Howell RR, Morriss FH, Pickering LK, editors. Human milk in infant nutrition and health, Illinois, USA: Charles C Thomas. 1986:3-L!?.

s24 P.M. Emmett, IS. Rogers I Early Humun Development 49 Suppl (1997) S7-S28

[2] Aggett PJ. Aspects of neonatal metabolism of trace metals. Acta Paediatr Stand 1994;402 Suppl:S75- S82.

[3] Aggett PJ, Hascheke F, Heine W, Hemell 0. Comment on the content and composition of lipids in infant formulas. Acta Paediatr Stand 1991;80:887-96.

[4] Ala-Houhala M. 25hydroxyvitamin D levels during breast-feeding with or without maternal or infantile supplementation of vitamin D. J Pediatr Gastroenterol Nutr 1985;4:220-6.

[5] Anderson DM, Pittard WB. Vitamin E and C concentrations in human milk with maternal megadosing: a case report. J Am Diet Assoc 1985;85:715-7.

[6] Arroyave G, Beghin I, Flores M et al. Efectos de1 consumo de azucar fortificada con retinol, en la madre embarazada y lactante cuya dieta habitual es baja en vitamina A. Estudio de la madre y de1 nine. Arch Latinoamericanos Nutr 1974;24:485.

[7] Balmer SE, Wharton BA. Diet and faecal flora in the newborn: iron. Arch Dis Child 1991;66:1390-4. [8] Belavady B, Gopalan C. Effect of dietary supplementation on the composition of breast milk. Ind J

Med Res 1960;48:518-23. [9] Brock JH. Lactoferrin in human milk: its role in iron absorption and protection against enteric

infection in the newborn infant. Arch Dis Child 1980;55:417-21. [lo] Brown K et al. Lactational capacity of marginally nourished mothers; Relationships between maternal

nutritional status and quantity and proximate composition of milk. Pediatrics 1986;78:909-19. [ll] Brown KH, Robertson AD, Akhtar NA. Lactational capacity of marginally nourished mothers:

infants’ milk nutrient consumption and patterns of growth. Pediatrics 1986;78:920-7. [12] Butte N et al. Variability of macronutrient concentrations in human milk. Eur 3 Clin Nutr

1988;42:345-9. [13] Carlson SE. Human milk non-protein nitrogen: occurrence and possible functions. Adv Pediatr

1985;32:43-63. [ 141 Chang S-J. Antimicrobial proteins of maternal and cord sera and human milk in relation to maternal

nutritional status. Am J Clin Nutr 1990;51:183-7. [15] Chappell JE, Francis T, Clandinin MT. Vitamin A and E content of human milk at early stages of

lactation. Early Hum Dev 1985; 11: 157-67. [16] Chappell JE, Francis T, Clandinin MT. Simultaneous high-performance chromatography analysis of

retinol esters and tocopherol isomers in human milk. Nutr Res 1986;6:849-52. [17] Coppa GV, Gabrielli 0, Giorgi P, Catassi C et al. Preliminary study of breastfeeding and bacterial

adhesion to uroepithelial cells. Lancet 1990;335:569-71. 1181 Coppa GV, Gabrielli 0, Pierani P, Catassi C, Carlucci A, Giorgi PL. Changes in carbohydrate

composition in human milk over 4 months of lactation. Pediatrics 1993;91:637-41. [19] Cumming FJ, Briggs MH. Changes in plasma vitamin A in lactating and non-lactating oral

contraceptive users. Br J Obstet Gynaecol 1983;90:73-7. [20] Dagnelie PC, van Staveren WA, Roos AH, Tuinstra LGMT, Burema J. Nutrients and contaminants in

human milk from mothers on macrobiotic and omnivorous diets. Eur J Clin Nutr 1992;46:355-66. [21] Davies DP. Plasma osmolality and feeding practices of healthy infants in first three months of life. Br

Med J 1973;2:340-2. [22] Deodhar AD, Ramakrishnan CV Relation between dietary intake of lactating women and the

chemical composition of the milk with regard to vitamin content. J Trop Pediatr 1960;6:44. [23] Department of Health and Social Security. The composition of mature human milk. London: HMSO,

1977. 1241 Department of Health and Social Security. Present day practice in infant feeding: Third report.

London:HMSO, 1988. [25] Dewey KG, Lovelady CA, Nommsen-Rivers LA, McCrory MA, Lonnerdal B. A randomized study of

the effects of aerobic exercise by lactating women on breast-milk volume and composition. New Engl J Med 1994;330:449-53.

[26] Donangelo CM, Trugo NMF, Koury JC, Barretosilva MI, Freitas LA, Feldheim W, Barth C. Iron, zinc, folate and vitamin B12 nutritional status and milk composition of low-income Brazilian mothers. Eur J Clin Nutr 1989;43:253-66.

[27] Donovan SM. Lonnerdal BO. Non-protein and true protein in infant formulas. Acta Paediatr Stand 1989;78:497-504.

P.M. Emmett, IS. Rogers I Early Human Development 49 Suppl (19971 57-328 S?.’

[28] Edidin DV, Levitsky LL, Schey W, Dumbovic N, Campos A. Resurgence of nutritional rickety associated with breast-feeding and special dietary practices. Pediatrics 1980$5:232-5.

1291 Edwards CA, Parrett AM, Balmer SE, Wharton BA. Faecal short chain fatty acids in breat-fed anti formula-fed babies. Acta Paediatr Stand 1994;83:459-62.

[30] Farquharson J, Cockbum F, Patrick WA, Jamieson EC, Logan RW. Infant cerebral cortex phos- pholipid fatty-acid composition and diet. Lancet 1992;340:810-3.

[31] Finley DA, Lonnerdal B, Dewey KG, Grivetti LE. Breast milk composition: fat content and fatty acid composition in vegetarians and non-vegetarians. Am J Clin Nutr 1985;41:787-800.

132) Foman SJ, Bier DM, Matthews DE, Rogers RR, Edwards BB, Ziegler EE, Nelson SE. Bioavailabiliry of dietary urea nitrogen in the breast-fed infant. J Pediatr 1988;113:515-7.

1331 Forsum E, Lonnerdal B. Effect of protein intake on protein and nitrogen composition of breast milk. Am J Clin Nutr 1980;33:1809-13.

1341 Friss HE, Rubin LG, Carsons S et al. Plasma fibronectin concentrations in breast fed and formula fed neonates. Arch Dis Child 1988;63:528-32.

1351 Funk MA, Hamlin L, Picciano MF, Prentice A, Milner JA. Milk selenium of rural African women influence of maternal nutrition, parity, and length of lactation. Am J Clin Nutr 1990;s 1:220-4.

1361 Galeano NF, Darling P, Lepage G, Leroy C, Collet S, Giguere R, Roy CC. Taurine supplementation of a premature formula improves fat absorption in premature infants. Pediatr Res 1987;22:&7--71.

[37] Garg M, Thirupuram S, Saha K. Colostrum composition, maternal diet and nutrition in North India. J Trop Pediatr 1988;34:79-87.

[38] Gasparoni A, Avanzini A, Probizer R et al. IgG subclasses compared in maternal and cord serum and breast milk. Arch Dis Child 1992;67:41-3.

[39] Gebre Mehdin M,Vahlquist B, Hofvander Y et al. Breast milk composition in Ethiopian and Swedish women. Am J Clin Nutr 1976;29:441.

[40] Golding J, Emmett PM, Rogers IS. Does breast feeding protect against non-gastric infections‘? Early Hum Dev 1997;49 Suppl:S105-S120.

1411 Golding J, Emmett PM, Rogers IS. Gastroenteritis, diarrhoea and breast feeding. Early Hum Dev 1997;49 suppl:s83-s1o3.

[42] Golding J, Rogers IS, Emmett PM. Association between breast feeding, child development and behaviour. Early Hum Dev 1997;49 Suppl:S175-S184.

[43] Gopalan C. Studies on lactation in poor Indian communities. J Trop Pediatr 1958;4:87-97. [44] Gopalan C, Belvady B. Nutrition and lactation. Fed Proc 1961;20(1 part 3). [4S] Greer FR, Marshall S, Cherry J, Suttie JW. Vitamin K status of lactating mothers. human milk and

breast-feeding infants. Pediatrics 1991;88:751-6. 146) Hachey DL, Silber GH, Wong WW. Human lactation II: Endogenous fatty acid synthesis by the

mammary gland. Pediatr Res 1989;25:63-8. [47] Hachey DL, Thomas MR, Emken EA. Human lactation: Maternal transfer of dietary triglycerides

labelled with stable isotopes. J Lipid Res 1987;28:1185-92. 1481 Harzer G, Dieterich I, Haug M. Effects of the diet on the composition of human milk. Ann Nutr

Metab 1984;28:23 l-9. [49] Hemell G. The requirements and utilization of dietary fatty acids in the newborn infant. Acta Paediatr

Stand, Suppl 1990:365:20-7. [SO] Hernell 0, Blackberg L. Digestion of human milk lipids: Physiologic significance of sn-2

Monoacylglycerol hydrolysis by bile salt-stimulated lipase. Pediatr Res 1982;16:882-5. [5 I ] lnsull W. Hersch J, James T, Ahrens EH. The fatty acids of human milk. Il. Alterations produced b)

manipulation of caloric balance and exchange of dietary fats. J Clin Invest 1959;38:443-50. 1521 Jackson AA, Golden MHN. The human rumen. Lancet 1978;1978:764-7. [53] Jelliffe DB. The protein content of the breast milk of African women. Br Med J 1953;2: 113 i -3. [54] Jelliffe DB, Jelliffe EFP. The volume and composition of human milk in poorly nourished

communities. A review. Am J Clin Nutr 1978;31:492-515. [55] Kang-Yoon SA, Kirksey A, Giacoia G, West K. Vitamin B-6 status of breast-fed neonates: influence

of pyridoxine supplementation on mothers and neonates. Am J Clin Nutr 1992;56:548-58. [S6] Karmarkar MG, Ramakrishnan CV. Relation between the dietary intake of lactating women and the

chemical composition of the milk with regard to principal and certain inorganic constituents. Acta Paediatr Stand 1960;49:599-604.

S26 P.M. Emmett, 13. Rogers I Early Human Development 49 Suppl (1997) S7-S28

[57] Karra MV, Kirksey A, Galal 0, Bassily NS, Harrison GG, Jerome NW. Effect of short-term oral zinc supplementation on the concentration of zinc in milk from American and Egyptian women. Nutr Res 1989;9:471-8.

[58] Kon SK, Mawson EH. Milk: The mammary gland and its secretion. Medical Research Council Special Report Series, 1950.

[59] Krebs NF, Hambridge KM, Jacobs MA, Rasbach JO. The effects of a dietary zinc supplement during lactation on longitudinal changes in maternal zinc status and milk zinc concentrations. Am J Clin Nutr 1985;41:560-70.

[60] Kumpulainen J, Vuori E, Makinen S et al. Dietary chromium intake of lactating Finnish mothers: effect on the chromium content of their breast milk. Br J Nutr 1980;44:257-63.

1611 Lamounier JA, Danelluzzi JC, Vannucchi H. Zinc concentrations in human milk during lactation: a 6-month longitudinal study in southern Brazil. J Trop Pediatr 1989;35:31-4.

[62] Lane PA, Hathaway WE. Vitamin K in infancy. J Pediatr 1985;106:351-9. [63] Laskey MA, Prentice A, Shaw J, Zachou T, Ceesay SM,Vasquez-Velasquez L, Fraser DR. breast-milk

calcium concentrations during prolonged lactation in British and rural Gambian mothers. Acta Paediatr Stand 1990;79:507-12.

[64] Lonnerdal B. Dietary factors affecting trace element bioavailability from human milk, cow’s milk and infant formulae. Prog Food Nutr Sci 1985;9:36-62.

[65] Lonnerdal B, Forsum E, Hambraeus L. A longitudinal study of the protein, nitrogen and lactose contents of human milk from Swedish well-nourished mothers. Am J Clin Nutr 1976;29:1127-33.

[66] Lucas A, Ewing G, Roberts SB, Coward WA. How much energy does the breast fed infant consume and expend?. Br Med J 1987;295:75-7.

[67] Lucas A, Sarson DL, Blackbum AM et al. Breast vs bottle: Endocrine responses are different with formula feeding. Lancet 1980;I: 1267-9.

[68] McClelland DBL, McGrath J, Samson RR. Antimicrobial factors in human miIk. Studies of concentration and transfer to the infant during the early stages of lactation. Acta Paediatr Stand 1978;271 Suppl:Sl-S20.

[69] Metz J, Zalusky R, Herbert V Folic acid binding by serum and milk. Am J Clin Nutr 1968;21:289- 97.

[70] Michaelsen KF, Larsen PS, Thomsen BL, Samuelsen G. The Copenhagen cohort study on infant nutrition and growth: breast-milk intake, human milk macronutrient content, and influencing factors. Am J Clin Nutr 1994;59:600-11.

[71] Millis J. The influence of nutrition on the growth rate in the first year of life of Chinese infants born in Singapore in 1951. Med J Austr 1954;1:322-7.

[72] Millis J. Modifications in food selection observed by Malay women during pregnancy and after confinement. Med J Malaya 1958;13:139-43.

[73] Moore MEC, Moran JR, Greene HL. Zinc supplementation in lactating women: Evidence for mammary control of zinc secretion. J Pediatr 1984;105:600-2.

[74] Moser PB, Reynolds RD. Dietary zinc intake and zinc concentrations of plasma erythrocytes and breastmilk in antepartum and postpartum lactating women: a longitudinal study. Am J Clin Nutr 1983;38:101-8.

1751 Moser PB, Reynolds RD, Acharya S, Howard MP, Andon MB. Calcium and magnesium dietary intakes and plasma and milk concentrations of Nepalese lactating women. Am J Clin Nutr 1988;47:735-9.

[76] Moser PB, Reynolds RD, Acharya S, Howard MP, Andon MB, Lewis SA. Copper, iron, zinc and selenium dietary intake and status of Nepalese lactating women and their breast-fed infants. Am J Clin Nutr 1988;47:729-34.

[77] Moser-Veillon PB, Reynolds RD. A longitudinal study of pyridoxine and zinc supplementation of lactating women. Am J Clin Nutr 1990;52:135-41.

[78] Motil KJ, Montandon CM, Hachey DL, Boutton TW, Klein PD, Garza C. Relationships among lactation performance, maternal diet, and body protein metabolism in humans. Eur J Clin Nutr 1989;18-37:681-91.

[79] Nagra SA. Longitudinal study in biochemical composition of human milk during first year of lactation. J Trop Pediatr 1989:35:126-8.

P.M. Emmett, IS. Rogers I Early Human Developmrnr 49 Suppl (1997) S7-328

[80] The National Academy of Sciences. Nutrition during lactation. Washington, D.C.: National Academy Press, 1991,

[Sl] Neville MC, Keller R, Seacat J, Lutes V, Neifert M, Casey C, Allen J, Archer P, Studies in human lactation: milk volume in lactating women during the onset of lactation and full lactation. Am J Clin Nutr 1988;48:1375-86.

[82] Ojofeitimi EO. Elegbe IA. The effect of early initiation of colostmm feeding on proliferation of intestinal bacteria in neonates. Clin Pediatr 1982;21:39-42.

[83] Pabst H, Spady DW. Effect of breastfeeding on antibody response to conjugate vaccine. Lancer 1990;336:269-70.

[84] Picciano MF. The volume and composition of human milk. In: Bond ST, Filer LJ, Leville GA, Thomson A, Weil WB, Infant and child feeding, editors. London: Academic Press, 1981:47-61.

[SS] Pietschnig B, Haschke F, Vanura H, Shearer M, Veitl V, Kellner S, Schuster E. Vitamin K in hreasr milk: no influence of maternal dietary intake. Eur J Clin Nutr 1993;47:209-15.

1861 Poskitt EME. Use of cows’ milk in infant feeding with emphasis on the compositional differences between human and cows’ milk. In: Walker AF, Rolls BA, editors. Infant nutrition. London: Chapman and Hall, 1994:163-185.

1871 Prentice A. Breast feeding increases concentrations of 1gA in infants’ urine. Arch Dis Child 1987;62:792-5.

[88] Prentice A, Barclay DV. Breast-milk calcium and phosphorus concentrations of mothers in rural Zaire. Eur J Clin Nutr 1991;45:611-7.

[89] Prentice A, Ewing G, Roberts SB, Lucas A, Mac&thy A, Jarjou LMA, Whitehead RG. Thr nutritional role of breast-milk IgA and lactoferrin. Acta Paediatr Stand 1987;76:592-8.

[90] Prentice AM, Lucas A, Vasquez-Velasquez L, Davies PSW. Whitehead RG. Are current dieta guidelines for young children a prescription for overfeeding?. Lancet 1988;ii: 1066-9.

[91] Prentice A, Prentice AM, Whitehead RG. Breast-milk fat concentrations of rural African women 1. Short-term variations within individuals. Br J Nutr 1981;45:483-94.

[92] Prentice AM, Roberts SB, Prentice A, Paul AA, Watkinson M, Watkinson AA, Whitehead RG. Dietary supplementation of lactating Gambian women. I. Effect on breast-milk volume and quality. Hum Nutr: Clin Nutr 1983;37(3:53-64.

[93] Rattigan S, Ghisalberti AV, Hartmann PE. Breast-milk production in Australian women. Br .I Nutr 1981;45:243-9.

[94] Rogers IS, Emmett PM, Golding J. The effects of lactation on the mother. Early Hum Dev 1997:49 Suppl:S 191 -S203.

19.51 Rothberg AD, Pettifor JM, Cohen DF, Sonnendecker EWW, Ross FP. Maternal-infant vitamin D relationships during breast-feeding. J Pediatr 1982;101:500-3.

[96] Saha K, Garg M, Rao KN, Thirupuram S, Gupta MM. Lymphocyte subsets in human colostrum with special reference to that of undernourished mothers. J Trop Pediatr 1987;33:329-32.

1971 Salariya EM, Robertson CM. Relationships between baby feeding types and patterns. gut transit time of meconium and the incidence of neonatal jaundice. Midwifery 1993;9:235-42.

[98] Salmenpera L, Perheentupa J, Nanto V, Siimes MA. Low zinc intake during exclusive breast-feeding does not impair growth. J Pediatr Gastroenterol Nutr 1994;18:361-70.

[99] Salmenpera L, Perheentupa J, Sumes MA. Folate nutrition is optimal in exclusively breast fed infants but inadequate in some of their mothers and in formula fed infants. J Pediatr Gastroenterol Nun 1986;5:283-9.

[IOO] Sanders THB, Ellis TR. Dickerson JWT. Studies of vegans: the fatty acid composition of plasma choline-phosphoglycerides, erythrocytes, adipose tissue, breastmilk and some indicators of suscep- tibility to ischemic heart disease in vegans and omnivore controls. Am J Clin Nutr 1978;31:805.

[ IOI] Scheifele D, Bjomson GJ, Guasparini R, Friesen B, Meekison W. Breastfeeding and antibody responses to routine vaccination in infants. Lancet 1992;340: 1406.

11021 Schreiner RL, Brady MS, Franken EA et al. Increased incidence of lactobezoars in low birth weigh! infants. Am J Dis Child 1979;133:936-40.

[ 1031 Shubat PJ, Parker D, Huxtable RJ. Effect of suckling and diurnal influences on the concentrations of taurine and other free amino acids in milk. Eur J Clin Nutr 1989;43:675-80.

S28 P.M. Emmett, IS. Rogers I Early Human Development 49 Suppl (1997) S7-$28

[ 1041 Smith AM, Picciano MF, Deering RH. Folate supplementation during lactation: maternal folate status and, human milk folate content, and their relationship to infant folate status. J Pediatr Gastroenterol Nutr 1983;2:622-8.

[105] Specker B. Vitamin B12: low milk concentrations are related to low serum concentrations in vegetarian women and to methylmalonic aciduria in their infants. Am J Clin Nutr 1990;52:1073-6.

[106] Specker BL, Valanis B, Hertzberg V, Edwards N, Tsang RC. Sunshine exposure and serum 25-hydroxyvitamin D concentrations in exclusively breast-fed infants. J Pediatr 1985;107:372-6.

[IO71 Squires BT. Ascorbic acid content of the milk of Tswana women. Royal Sot Trop Med Hyg 1950;53:170.

[108] Stephens S. Development of secretory immunity in breast fed and bottle fed infants. Arch Dis Child 1986;61:263-9.

[109] Stoltzfus RJ, Hakimi M, Miller K, Rasmussen KM, Dawiesab S, Habicht J, Dibley M. High dose vitamin A supplementation of breastfeeding Indonesian mothers: Effects on the vitamin A status of mother and infant. J Nutr 1993;123:666-75.

[ 1 lo] Tamura T, Yoshimura Y, Arakawa T. Human milk folate and folate status in lactating mothers and their infants. Am J Clin Nutr 1980;33:193-7.

[ill] The Royal Society of Chemistry and MAFF. Milk products and eggs. London: HMSO, 1989. [112] The Royal Society of Chemistry and MAFF. The composition of foods. London: HMSO, 1991. [ 1131 Van Woelderen BF. Changing insights into human milk proteins: some implications. Nutr Abstr Rev

(Series A) 1987;57:129-33. [114] Villard L, Bates CJ. Effect of vitamin A supplementation on plasma and breast milk vitamin A

levels in poorly nourished Gambian women. Hum Nutr: Clin Nutr 1987;41C:47-58. [115] Walker WA. Absorption of protein and protein fragments in the developing intestine: role in

immunologic/allergic reactions. Pediatrics 1985;75 Suppl:S167-71. [116] Wasserman RH. Lactose stimulates gastrointestinal absorption of calcium: theory. Nature

1964;139:997-9. [117] Wheeler E, Tan SP. From concept to practice: Food behaviour of Chinese immigrants in London,

Ecol Food Nutr 1983;13:51-7. [118] Whitehead RG, Paul AA. Growth charts and the assessment of infant feeding practices in the

western world and in developing countries. Early Hum Dev 1984;9:187-207.

Properties of human milk and their relationship with maternal nutrition

Documents