Food and microbiological problems in the newborn: data and practice.

We have reviewed the relationships of food, nutrition and feeding practices to various infections in the newborn. Tentative conclusions are made: (a) the initial use of human milk (raw or pasteurized) continues to offer advantages in the care of babies in intensive care; (b) attempts to mimic the microbiological effects of breast milk by manipulation of the composition of infant formulas have so far achieved little success, but this is a rapidly developing field; (c) we are wary of the widespread use of breast milk "fortifiers" until there is evidence that they do not adversely affect the protective properties of breast milk; (d) the doubtful advantages of nasojejunal feeding need to be weighed against the increased bacterial contamination of the upper small bowel; (e) systems monitoring in milk kitchens and the handling of feeds in the neonatal unit are an integral part of comprehensive neonatal care; (f) to limit nosocomial infection, particular attention to the faecal-food-oral route is necessary since there is potential for multiplication of initial contamination of food.

Sorrento studies of diet and fecal flora in the newborn.

The fecal flora of a breast-fed baby is very different from that of a bottle-fed baby. This paper reviews five previous studies, performed at this hospital concerning the effect of various dietary components (whey proteins, casein, lactoferrin, iron, nucleotides) on the fecal flora. The babies received either breast milk or one of the test formulas from birth. Fecal samples were examined by quantitative microbiological methods at 4 and 14 days and at various intervals thereafter. By 14 days differences in the fecal flora were established. Among breast-fed babies bifidobacteria, lactobacilli and staphylococci were predominant organisms, whereas in the formula-fed babies the predominant organisms were enterococci, coliforms, and Bacteroides. A whey-based formula without bovine lactoferrin, iron or nucleotides gave a flora a little closer to but still remote from the breast-fed one. Despite extensive modification of cow's milk in the manufacture of a modern infant formula, the fecal flora of bottle-fed babies remains substantially different from that of breast-fed babies.

Protein nutrition, faecal flora and iron metabolism: the role of milk-based formulae.

This paper explores the role of milk-based formulae in achieving four aspects of nutritional health in infants and toddlers: in the suckling, to mimic the amino acid metabolism and the faecal flora of a breast-fed baby; in the weanling, to achieve adequate protein intakes in later infancy and beyond and to achieve satisfactory haemoglobin concentrations in the early toddler years. Milk-based formulae have two roles in infant nutrition: as so-called breast milk substitutes and as a safety net during the weaning period; the latter role may be the more important.

Faecal short chain fatty acids in breast-fed and formula-fed babies.

The intestinal flora of breast-fed infants differs from that of formula-fed infants. It is thought that this difference in flora may be one important reason why breast-fed babies suffer less from gastrointestinal disease. Differences in intestinal flora are reflected in the profile of faecal short chain fatty acids (SCFA). Very little is known about faecal concentrations of SCFA in babies fed breast milk or infant formula. In this study, faecal SCFA were measured in babies at two and four weeks of age who had been either exclusively breast fed or bottle fed from birth. There was no significant difference in total faecal SCFA concentrations between breast-fed and formula-fed babies when lactate was included. The formula-fed group, however, had less lactic acid and higher concentrations of propionic and n-butyric acids than breast-fed babies. Very few babies had significant levels of n-butyric acid, although this SCFA is believed to be important for the health of the colonic mucosa of adults.

Diet and faecal flora in the newborn: nucleotides.

Breast milk contains nucleotide salts that are only present in minimal amounts in modern infant formulas prepared from cows' milk. Nucleotides have been suggested as cofactors for the growth of bifidobacteria in vitro. Bifidobacteria are found to be more numerous in the faeces of breast fed babies compared with those of formula fed babies. Faecal flora were examined at 2 weeks of age in 32 babies who from birth had been fed a whey based formula supplemented with nucleotide monophosphate salts, 33 babies fed an unsupplemented formula, and 21 breast fed babies. Faecal flora were also examined at 4 weeks, and 7 weeks but with fewer babies in each group. Most differences were found at 2 weeks of age when more babies fed the nucleotide supplemented formula were colonised with Escherichia coli and more had E coli as the dominant organism in their faecal flora. Fewer of these babies were colonised with bifidobacteria. The counts of bifidobacteria and enterococci were reduced in the nucleotide supplemented group but bacteroides accounted for a higher percentage of the total flora in this group of babies. Supplementation of a formula with nucleotide salts did not make the faecal flora closer to that of breast fed infants as the growth of bifidobacteria was discouraged. While there may be arguments to support the addition of nucleotides to infant formula the results of this study do not support their addition for the enhancement of bifidobacteria in the faecal flora.

Faecal flora in the newborn. Effect of lactoferrin and related nutrients.

Bifidobacteria, lactobacilli and staphylococci are the predominant organisms in the faeces of breast fed babies whereas in formula fed babies coliforms, enterococci and bacteroides predominate. In vitro studies suggest that the mechanisms responsible are probably related to the acid base properties of the formula and 'immunological' proteins such as lactoferrin and sIgA. In human babies however the addition of bovine lactoferrin to an infant formula has little effect on the faecal flora and does not move it in the direction of the breast fed baby. There are various possible explanations of this lack of effect, e.g., inactivation of the lactoferrin when it is added to a formula, and immunological responses in the intestine to a foreign protein. We consider the most likely explanation is that other factors necessary for the optimum activity of lactoferrin were not present or in inappropriate concentration, e.g. sIgA, lysozyme, citrate, bicarbonate. If human lactoferrin is added to an infant formula it may be these other factors will require attention if the lactoferrin is to have a significant effect. An iconoclastic interpretation which cannot be completely excluded is that the hypothesis of lactoferrin bacteriostatic activity is based on in vitro studies and is not a reflection of what happens in babies. The faecal flora of a breast fed baby is very different from that of a baby receiving either cow's milk or a modern infant formula (Figure 1). Among breast fed babies bifidobacteria lactobacilli and staphylococci are the predominant organisms, whereas in formula fed babies the predominant organisms are enterococci, coliforms, and bacteroides (1).

Diet and faecal flora in the newborn: iron.

The faecal flora of a baby receiving a modern infant formula is substantially different from that of a breast fed baby. This difference is a little less when whey based formulas are used. The addition of bovine lactoferrin has no effect and there is some evidence that the presence of added iron in a formula moved the faecal flora further away from that of a breast fed baby. the iron content of currently used infant formulas is much higher than that of breast milk. The effect of the addition of iron to both whey and casein based formulas on the faecal flora was examined in further detail. Faecal flora were examined at 14 days of age in 33 babies receiving a whey formula fortified with iron, 29 babies receiving a whey formula without iron, 29 babies receiving a casein formula fortified with iron, and 24 babies receiving casein formula without iron. Subsequently fewer babies in each group were examined at week 7, 11, and 15. The addition of iron to both casein and whey formulas discouraged colonisation and growth of staphylococci and bacteroids but encouraged the colonisation and growth of clostridia and enterococci. The type of protein and not the iron content had more effect upon the growth of bifidobacteria; both whey formulas, fortified or not, encouraged the colonisation by bifidobacteria. If an infant formula, for use in the first few months, is to mimic the physiological effects of breast milk, there may be microbiological arguments for not fortifying it with iron. However, large empirical trials would be necessary before advocating such a policy.

Diet-related reference values for plasma amino acids in newborns measured by reversed-phase HPLC.

We have measured by reversed-phase HPLC concentrations of amino acids in plasma in groups of 80 normal appropriate-weight term babies fed from birth either a casein formula (WhiteCap SMA, n = 26), a whey formula (Gold Cap SMA, n = 26), or breast milk (n = 28). They were studied from day 11 to week 15 postpartum. The trend was towards an increase in amino acid concentrations in plasma with age, more marked in formula-fed than in breast-fed infants. Reference values were derived for each group. Both formula-fed groups showed several differences from the breast-fed group. Detailed examination indicated that tyrosine, phenylalanine, and methionine concentrations were increased in the casein-fed group greater than 20% of the time, but only threonine was similarly increased in the whey-fed group. Other amino acids, different ones for each formula group, were increased less frequently. There were no consistent correlations with any aspect of infant growth. Appropriate reference values are important for interpreting amino acid concentrations in plasma from newborns and for evaluating the effects of any future dietary modifications to infant formulas. HPLC analysis provides a suitable highly sensitive method for undertaking such studies.

Diet and faecal flora in the newborn: lactoferrin.

The faecal flora of breast fed babies differs from that of bottle fed babies. We have shown that the use of a whey predominant formula rather than a casein predominant one induced a faecal flora generally closer to that of breast fed babies but substantial differences remained. The whey proteins of breast milk include much more lactoferrin than is found in cows' milk. Observations both in animals and in vitro suggest that lactoferrin could be responsible for some of these differences between bottle and breast fed babies. This study was designed to determine the effects on faecal flora of the addition of bovine lactoferrin to the diet of bottle fed babies while holding other qualities of their diet constant. As lactoferrin is an iron binding protein three test formulas were used: (a) no added iron and no added lactoferrin (basic), (b) no iron but added lactoferrin (L), and (c) added iron and lactoferrin (LF). The addition of lactoferrin had little effect upon the faecal microflora and did not move the pattern of the faecal flora in the direction of the breast fed baby. The addition of iron to the formula had more effect on the faecal flora than did lactoferrin. At day 4 it encouraged Escherichia coli and discouraged staphylococcal faecal colonisation. At day 14 the addition of iron to the formula discouraged bifidobacteria. The reasons why bovine lactoferrin was ineffective in vivo in this study are discussed.

Diet and faecal flora in the newborn: breast milk and infant formula.

This study examined the faecal flora on days 4, 14, and 28 of 17 breast fed babies and 26 bottle fed babies receiving a modern infant formula based on demineralized whey. Generally among breast fed babies bifidobacteria and staphylococci were the predominant organisms, whereas in the formula fed babies the predominant organisms were enterococci, coliforms, and clostridia. Despite the extensive modification of cows' milk to make an infant formula resemble human breast milk, the results are very similar to those previously reported with unmodified cows' milk baby feeds. The exact dietary factor responsible for these microbiological differences is unclear and in succeeding papers we have looked at the effects of protein quality, in particular the content of whey proteins, casein, and lactoferrin.

Diet and faecal flora in the newborn: casein and whey proteins.

Despite the extensive modifications of cows' milk to make an infant formula resemble human breast milk, we showed in a previous study that the faecal flora of breast fed babies still differs substantially from that of formula fed babies. This paper describes the effects that differences in the distribution of whey proteins and caseins exert on the faecal flora. Faecal flora was examined in 33 babies receiving a whey formula, 29 babies receiving a casein formula, and 38 breast fed babies. Subsequently fewer babies in each group were studied at weeks 7, 11, and 15. More whey fed babies were colonised with bifidobacteria at 14 days compared with the casein fed group, more casein fed babies were colonised with bacteroides (at 14 days), and more had a dominant growth of enterococci (at week 7). It seems therefore that during the first two months that the whey predominant formula induced a faecal flora generally closer to that of breast fed babies than did a casein formula.

Lactoferrin and iron absorption in newborn infants.

Results from experiments in this laboratory using 59Fe suggest that bovine lactoferrin (Lf) has no effect on iron absorption in rats. A study was therefore carried out in newborn infants to measure the effects of Lf on iron retention. Bovine Lf was labeled with the stable isotope 58Fe and fed to 7-day-old infants in a standard milk formula. Iron retention was estimated by measuring the unabsorbed 58Fe excreted in the feces during the following 3 days using neutron activation analysis. The results were compared with those obtained from a group of infants fed a similar level of iron as ferric chloride, labeled with 58Fe, together with 30 mg ascorbic acid. There was a very wide variation in percent iron retention amongst the infants but no overall difference between the Lf and ferric chloride groups. This confirms the previous findings in rats that Lf does not influence the availability of nonheme iron.