Intermediary metabolism

Throughout fetal life there is a progressively changing picture of enzymatic differentiation ( ). Certain enzymes of amino acid metabolism develop late, including those concerned with the synthesis of cysteine from methionine, taurine from cysteine and tyrosine from phenylalanine, with degradation of tyrosine and production of urea ( ). As a result, low-birthweight (LBW) infants might be expected to have increased dietary requirements for certain amino acids (such as cysteine and taurine; and be at risk for possibly deleterious accumulation of others (such as phenylalanine, tyrosine and methionine; Ch. 34.3 ).

Key enzymes in gluconeogenic pathways (e.g. phosphoenolpyruvate carboxykinase) may not develop until near or even just after term delivery ( ). A constant transplacental glucose infusion renders gluconeogenesis relatively unimportant in utero, and the fetal liver is more concerned with the storage of glucose as glycogen; phosphorylase and glucose 6-phosphatase ensure immediate glucose release after birth and defer the need for gluconeogenesis until around 24–48 hours of age; in contrast, the preterm neonate, born with low stores of liver glycogen ( ) and reduced gluconeogenic ability, is at risk of hypoglycaemia ( Ch. 34.1 ).

Gastrointestinal tract

See Chapter 29 .

Digestion and absorption

Luminal hydrolysis results in the breakdown of most large-molecular-weight proteins into peptides and amino acids. Most peptides are then hydrolysed by peptidases on the microvillus membrane prior to transport through the membrane, but some are absorbed intact. Although pepsin secretion is lower in preterm than term infants at birth, it is unaffected by the type of diet ( ). The activity of brush-border peptidases is also low in preterm infants at birth, but increases rapidly.

Protein requirements for term babies

These are discussed further in Chapter 16, part 2 . The EC Directive guidelines recommend that formulas containing unmodified cow’s milk protein should contain 1.1–2.1 g of protein per 100 ml of reconstituted feed. As growth velocity falls during infancy, there is a progressively decreased need for protein intake per kilogram of bodyweight. In recent years, there has been increasing speculation over the role of excess infant protein intake and increased risk of later obesity ( ); hence there have been moves to lower the protein intakes of term infants.

Estimating the protein requirements for the preterm baby

That the rapid growth in preterm babies might greatly increase the need for dietary protein has been appreciated for over 40 years. The increasing survival of extremely preterm and extremely low-birthweight (ELBW) infants, and the appreciation that these infants often acquire substantial nutrient deficits accompanied by compromised growth during early postnatal life, has led to a re-evaluation of their protein requirements. Recent recommendations take account of a number of factors and serve as a model for the investigation of requirements for other nutrients.

Protein quality

All protein supplies are not equal and it is recognised that net protein utilisation (N retained/N intake) differs with different feeding regimens. Whey proteins have a lower concentration of aromatic amino acids (tyrosine and phenylalanine) than are found in caseins; the studies by were performed with high-casein formulas (like cow’s milk), but most modern formulas are whey-predominant, reducing the possibility of hypertyrosinaemia and hyperphenylalaninaemia. Whey is also a good source of cysteine, a potentially essential amino acid in the newborn period (see above). However, there are still considerable differences in the protein composition of modern infant formulas compared with breast milk. A large part of this difference relates to the lower concentration of α-lactalbumin and higher concentration of β-lactoglubulin in bovine milk. It is now possible to make whey fractions from bovine milk with lower β-lactoglubulin and higher α-lactalbumin. These fractions contain higher concentrations of essential amino acids, particularly tryptophan and cysteine. Their addition to infant formulas may result in plasma amino acid patterns closer to those seen in breastfed infants, and allow a reduction in the total protein levels, with potential benefits for the infant’s growth pattern ( ; ).

Amino acid composition

There is increasing focus on amino acid requirements rather than protein requirements, on the basis that infants require specific amino acids, not proteins. Little is known about the optimal intakes of specific amino acids. However, a different amino acid composition of protein may well alter the total amount required. Breastfed infants have higher plasma and urine taurine concentrations and a higher rate of synthesis of bile acids than those fed on formula; the latter may partially explain the better fat absorption of infants fed human milk rather than formulas ( ). Rhesus monkeys fed a taurine-deficient formula for the first 6 or 12 months of life show abnormal retinal structure, although the abnormalities show some degree of spontaneous regression by 12 months even when the animals remain on the deficient diet ( ). A randomised trial of taurine supplementation in formula-fed preterm infants ( ) showed no effect on growth, behaviour or electroretinograms, but some evidence of more rapid auditory maturation in the supplemented group at the equivalent of term (as assessed by brainstem-evoked response). More recently, have shown that preterm infants with the highest plasma taurine concentrations during the neonatal period have better numeracy skills during adolescence and better development of the related parietal cortical grey matter. Thus, it seems prudent to add taurine to LBW formulas to achieve concentrations similar to those of breast milk.

Glutamine is used as a fuel by the small intestine and as a precursor for the synthesis of purine and pyrimidine bases. Numerous studies in animals and adult humans have demonstrated that it has beneficial effects on the GI tract, including the maintenance of structure and function during parenteral nutrition. A systematic review that included data from 2365 preterm infants who participated in seven randomised trials of either enteral or parenteral glutamine supplementation concluded there was no evidence of a beneficial effect of supplementation on mortality or short-term morbidity. The single study that assessed longer term outcomes found no effect on neurodevelopment at 18 months’ corrected age ( ).

Arginine is a precursor for the synthesis of nitric oxide, which in turn is important as a regulator of vascular perfusion. Plasma arginine concentrations have been found to be inversely related to the severity of respiratory distress syndrome, and low concentrations have been reported in infants who develop NEC. A single randomised trial of arginine supplementation (1.5 mmol/kg/day, given either enterally or parenterally) versus placebo in preterm infants found a significantly reduced incidence of NEC in supplemented infants (RR 0.24; 95% CI 0.10–0.61), with no adverse effects. However, given the high incidence of NEC in the study population, the data are insufficient at present to support a practice recommendation. A multicentre randomised controlled study of arginine supplementation in preterm neonates is needed, focusing on the incidence of NEC, particularly the more severe stages (2 or 3).

Digestion and absorption

Fat provides about half the energy for infants fed human milk, and its digestion commences in the stomach, catalysed by lingual lipase and gastric lipase ( ). Gastric lipolysis is quantitatively greater in the infant than in the adult, and the output and activity of lipases in the preterm infant are equal to those of adults maintained on a high-fat diet. Although gastric function and the production of lipase are unaffected by infant diet, the extent of fat digestion is greater in babies fed human milk (25%) than formula (14%), probably because of the structural differences between triglyceride in human milk fat globules and that in formula fat particles. Both lingual and gastric lipases are able to penetrate the milk fat globule membrane and digest triglyceride without disrupting its structure. The contribution of pancreatic lipases is relatively lower in infants than in adults. However, the bile salt-stimulated lipase (BSSL) present in human milk may contribute significantly to fat digestion. BSSL is present in high quantities even in the milk of mothers who deliver prematurely, and its concentration is independent of milk volume. Unlike gastric and pancreatic lipases, BSSL shows no positional or fatty acid specificity and is able to produce free fatty acids which are more easily absorbed than mono- or diglycerides at the low bile salt concentration seen in newborn infants.

The products of fat digestion are absorbed, resynthesised as triglycerides and secreted mainly into the lymphatic system as chylomicrons, and thence into the blood via the great veins. The foregoing applies to long-chain triglycerides, which are best absorbed if they are unsaturated. In contrast, medium-chain triglycerides (MCTs: 8–10 carbon atoms to the chain) are handled quite differently: their digestion is largely independent of bile salts; they are well absorbed, hydrolysed or intact, and pass to the liver via the portal system. Faecal fat excretion in newborn infants is greater in infants fed on cow’s milk than in those fed on human milk or vegetable fats.

Most modern formulas contain a mixture of animal and vegetable oils, adjusted to mimic the pattern of fatty acid saturation and chain lengths found in breast milk. When compared with human milk, such fat mixtures have a reduced content of fatty acids esterified to glycerol in the 2 position and an increase in those esterified in the 1 and 3 positions. The latter undergo hydrolysis in the gut, releasing palmitic acid, which is poorly absorbed and tends to form calcium soaps; this may be partly responsible for the harder stools seen in formula-fed infants, and could influence calcium absorption. A recent review identified nine publications in preterm or term infants in which ‘standard’ infant formulas containing high palmitate concentrations were compared with either low palmitate formulas or formulas containing a modified synthetic fat blend (Betapol) with a high proportion of fatty acids esterified in the 2 position to mimic that found in human milk ( ). Standardised results from these studies were consistently positive in favour of the low/synthetic palmitate groups with respect to intestinal fractional absorption of fat, palmitic acid and calcium. Total body bone mass was also significantly higher in two studies ( ; ), suggesting that increased calcium absorption results in measurable biological effects in the short term ( ). However, follow-up of one of these cohorts at age 10 years found no persisting effect of the intervention on bone mass, suggesting that any effect may be transient (Fewtrell et al. unpublished observations).

Fat requirements

Because there are clinical and physiological ceilings on the amount of dietary energy that it is desirable to supply as carbohydrate or protein, a minimum of around 30% of dietary energy needs to be supplied as fat. Linoleic and alpha-linolenic acids are essential fatty acids for the development of the brain and for prostaglandin synthesis. Essential fatty acid deficiency is also associated with skin lesions and retarded growth.

Two other dietary factors are important for lipid metabolism. Carnitine plays a key role by facilitating transport of long-chain fatty acids across the mitochondrial membrane prior to their oxidation ( ). Carnitine deficiency during the neonatal period has been reported in infants who experience intrapartum hypoxia and acidosis ( ). Preterm and small-for-gestational age (SGA) infants may have impaired endogenous carnitine synthesis, and if carnitine intake is deficient (as in total parenteral nutrition (TPN)) plasma and tissue concentrations fall. Nevertheless, there is currently insufficient evidence to judge whether such infants are put at clinical risk from a low-carnitine diet is uncertain. A randomised trial found that preterm infants who received supplemental carnitine in parenteral nutrition did not demonstrate any reduction in apnoea of prematurity, ventilatory requirements or need for supplemental oxygen therapy compared with those who received placebo ( ). Standard formulas usually contain similar concentrations of carnitine to those in breast milk, but some preterm formulas have additional carnitine.

Choline is required for phospholipid and acetylcholine synthesis ( ). About half the choline requirement is derived from the diet. Human and cow’s milk-based diets provide a sufficient intake.

Inositol, a six-carbon cyclic polyalcohol sugar, is a component of membrane phospholipids, and compounds containing inositol are important in signal transduction. Breast milk, particularly colostrum, contains high concentrations, whereas the levels in infant formulas are lower and intravenous feeding solutions have none. At present, it is recommended that all preterm infants receive supplementation based upon the level of inositol in human milk. A systematic review of three RCTs of inositol supplementation in preterm infants concluded that supplementation results in a statistically significant and clinically important reduction in the risk of chronic lung disease, retinopathy of prematurity (stage 4 or needing therapy), intraventricular haemorrhage (grade III–IV) and death ( ). The authors suggest a multicentre RCT of appropriate size is required to confirm these findings.

Essential fatty acid requirements

Two groups of long-chain polyunsaturated fatty acids (LCPUFAs, i.e. polyunsaturated fatty acids with greater than 18-carbon chain length) have received increasing interest in recent years: these are homologues of linoleic acid of the n-6 series (dihomogammalinolenic acid, arachidonic acid) and of α-linolenic acid of the n-3 series (eicosapentaenoic acid, docosahexaenoic acid (DHA)). The LCPUFAs are synthesised from the precursor essential fatty acids by a process of chain elongation and desaturation ( Fig. 16.1 ). They are found in high concentrations in the phospholipids of cell membranes, notably in the central nervous system ( ). In addition, arachidonic acid, dihomogammalinolenic acid and eicosapentaenoic acid are precursors for eicosanoids – important modulators and mediators of a variety of essential biological processes.

Major steps in the formation of long-chain polyunsaturated fatty acids (LPCs) from the C:18 essential fatty acids. Unsaturated fatty acids contain at least one double bond between adjacent carbon atoms: the number of double bonds is represented by 1n, 2n, 3n, etc. LPCs are further classified by the position of the first double bond from the methyl or omega end of the hydrocarbon chain, represented as -3, -6, -9 or _-3, _-6, _-9. Linoleic acid (C18:2n-6) thus has a chain length of 18 carbons with two double bonds, the first of which is at the sixth carbon atom from the omega end.

Rapid accumulation of LCPUFAs in the brain, particularly DHA, occurs from the third trimester to 18 months postpartum ( ). Human milk contains both the precursor essential fatty acids and adequate LCPUFAs for structural lipid accretion ( ), but infant formulas traditionally contained only the parent essential fatty acids, the assumption being that the infant could synthesise LCPUFAs from these. However, term and preterm infants fed on formulas which contain minimal LCPUFAs have been shown to have lower red cell LCPUFAs and lower LCPUFAs in the phospholipids of the cerebral cortex and subcutaneous tissues than infants fed breast milk ( ; ; ). Largely on the basis of the biochemical data, recommended several years ago that formulas for LBW infants should be enriched with metabolites of both linoleic and linolenic acids, approximating the levels typical of human milk; indeed, the majority of preterm and term infant formulas now contain preformed LCPUFAs. However, although there is clear evidence that supplementing infant formulas with LCPUFAs results in biochemical improvement, there is less evidence of lasting clinical benefits; recently updated systematic reviews concluded there was no evidence for a clinical benefit of LCPUFA supplementation of term ( ) or preterm ( ) infant formula. In the most recent and largest trial conducted in this field, preterm infants (<33 weeks’ gestation) were randomised to receive either high DHA (1% of total fatty acids) versus a typical current DHA intake (0.3%) achieved by a combination of supplementing mothers who were expressing milk for their infant and the use of DHA-supplemented formulas. No effect of high-dose supplementation was seen overall at 18 ( ) or 26 months’ ( ) corrected age, although girls had significantly higher Bayley MDI at 18 months. Longer term follow-up studies are currently in progress, testing for effects of LCPUFAs on health outcomes, including cardiovascular risk as well as cognitive function.

Term babies

The EC Directive guidelines ( ) recommend that the fat content of infant formulas should lie between 2.6 and 4.1 g per 100 ml of feed. The guidelines also specify levels for alpha-linoleic acid (not less than 12 mg/100 kcal) and linoleic acid (300–1200 mg/100 kcal), with the linoleic/alpha-linoleic ratio between 5 and 15, for term infant formulas. However, there is currently no requirement to add LCPUFAs, although, if added, their content shall not exceed 1% of total fat for n-3 LCP and 2% of total fat for n-6 LCP. Furthermore, the DHA content shall not exceed that of the n-6 LCP and the eicosapentaenoic acid content shall not exceed the DHA content.

Preterm babies

The main problem with dietary fat in preterm infants is the increased tendency to steatorrhoea. Reduced fat absorption in LBW infants relates to:

• •
  

  reduced pancreatic lipase and carboxylic ester hydrolase activity

  
  
• •
  

  reduced bile acid pool size and secretion rate: the duodenal bile acid concentration may well be below the critical level for micelle formation

  
  
• •
  

  possible reduction in activity or excretion of lingual lipase.

Fat absorption from fresh breast milk is approximately 90%, but the observed range is enormous. found that fat absorption from expressed breast milk fell to 55% in VLBW infants fed pasteurised milk, and to around 45% when the milk was boiled. These data may reflect loss of the bile salt-stimulated lipase found in human milk due to heat treatment.

A controversial issue is the addition of large quantities of MCTs to specially designed preterm infant formulas, largely because such babies absorb palmitic acid (n-16) poorly. The MCT content of human milk is low (less than 2% total fatty acids), whereas modern preterm formulas may contain up to 40% of the fat in this form. MCTs may spare dietary nitrogen and enhance calcium and magnesium absorption ( ). However, a review that included eight small randomised trials of MCT versus long-chain triglycerides in preterm infants concluded there was no evidence for an effect of MCT on short-term growth, GI intolerance, or NEC, although larger studies would be required to confirm these findings and to examine longer term outcomes ( ).

Physiology

The carbohydrate in human milk is almost entirely lactose, which provides 40% of ingested energy ( ). Other carbohydrates, e.g. sucrose and maltodextrins, may be hydrolysed efficiently by active brush-border sucrase, maltase and isomaltase, even in preterm infants, and starch or glucose polymers can be digested by salivary and pancreatic amylase, by amylase present in human milk, and by intestinal mucosal hydrolases.

Dietary lactose undergoes one of two processes: hydrolysis into glucose and galactose by the intestinal brush-border lactase, followed by absorption of glucose and galactose, or fermentation in the colon, with production of various gases and short-chain fatty acids. Energy from the latter can be absorbed, compensating at least in part for the inefficiency of dietary energy utilisation. Short-chain fatty acids administered into the colon have also been shown to stimulate intestinal growth following gut resection in animal models, and prevent mucosal atrophy after resection and TPN. There is controversy over the possible role of lactose fermentation in the development of NEC. However, although there are some experimental data linking excessive carbohydrate fermentation in the small intestine with an inflammatory condition resembling NEC, there is little evidence that colonic lactose fermentation is a primary factor.

Galactose and glucose are absorbed by the same carrier mechanism, and more than 90% reaches the portal vein. Most galactose is removed by the liver on first pass, and appears to be preferentially used for glycogen synthesis rather than conversion to glucose.

There is currently intense interest in the prebiotic role of oligosaccharides in human milk. Prebiotics are defined as non-digestible food ingredients that selectively stimulate the growth and/or activity of one or more bacteria in the colon, and therefore benefit the host. They promote the bifidogenic-dominant colonic microflora observed in breastfed infants, which in turn protects the infant from enteropathogenic bacteria. Human milk contains hundreds, if not thousands, of oligosaccharides; the composition is genetically determined with wide interindividual variation, making it difficult to define (and therefore to mimic) the oligosaccharide content of breast milk. Cow’s milk formulas traditionally contained no oligosaccharides, but, in recent years, oligosaccharide mixtures such as GosFos (a mixture of 90% galacto-oligosaccharides with 10% fructo-oligosaccharides) have been added. Randomised trials of supplementation of formula with oligosaccharides have shown enhanced colonisation with beneficial intestinal flora in both term and preterm infants, but there is less evidence for effects on clinical outcome ( ).

Term infants’ requirements

The EC Directive recommendation ( ) for carbohydrate intake is that it should be between 5.5 and 10.0 g per 100 ml of reconstituted feed, and that the lactose content should be above 2.7 g/100 ml ( Table 16.1 ). Other carbohydrate sources that are permitted in modern formulas are maltose, sucrose, glucose, maltodextrin, glucose syrup and precooked or gelatinised starch.

Preterm infants’ requirements

It is difficult to infer from physiological studies which carbohydrate is optimal for LBW infants. Lactose may enhance gut absorption of calcium and magnesium and may encourage a favourable gut flora; excessive intakes may result in diarrhoea and metabolic acidosis, yet in practice a high lactose intake in preterm infants is usually well tolerated.

One approach in the design of preterm formulas is to use lactose as the principal carbohydrate source, but to replace a proportion of the carbohydrate with glucose polymers in order to prevent excess osmolality (see Table 16.2 for recommended intakes).

Term infants’ requirements

The EC Directive guidelines recommend that formula should contain similar energy contents to those reported in human milk, i.e. 60–71 kcal/100 ml of reconstituted feed. However, from studies employing the doubly labelled water method, suggested that infants fed on human milk may receive lower energy intakes than those commonly reported and those employed in modern formulas – they obtained a value of 58 kcal/100 ml at 3 months; this may explain why modern formula-fed term infants grow faster than their breastfed counterparts.

Term SGA infants have been reported to have energy expenditures 5–10% higher than infants who are appropriate for gestational age, and may therefore benefit from increased energy to promote catch-up growth. In one study, a high-energy formula resulted in a marginal increase in weight gain and head growth ( ). However, it is likely that both extra energy and protein are required simultaneously.

Preterm infants’ requirements

The energy requirements for preterm infants cannot be calculated without some consideration of the desired composition of weight gain, as the deposition of different types of tissue incurs different energy costs. For example, 9 kcal are stored in each gram of fat, compared with 4 kcal per gram of protein. Thus, a weight gain high in fat will require more energy to be stored than one high in protein. As previously mentioned, the ratio of energy to protein is important in determining the relative amount of lean tissue versus fat which is deposited. These considerations are not just theoretical, as the different diets used for preterm infants do result in variations in the composition of tissue deposited ( Fig. 16.2 ). At present it is not known whether it is best to aim for a weight gain with 15% fat (as in the fetus) or nearer 40% fat (as in a term infant), and, unless this can be shown to have implications for future health, one might argue that it is academic. The results to date have been conflicting: and both found no differences in growth at 2 or 3 years of age in preterm infants who had shown wide differences in adiposity as infants. However, reported that greater adiposity at birth was a predictor of greater fatness at 6 years of age. In our own randomised trial of diet in preterm infants we found major differences in growth rates between diet groups during the neonatal period, but no differences in either growth ( ) or body fat mass or lean mass ( ) between diet groups at follow-up in childhood or early adult life ( ).

Weight gain composition of the fetus (32–36 weeks) and of the term infant (0–4 months), compared with that of preterm infants fed different diets. (A) Protein-supplemented pooled human milk. (B) Pooled human milk. (C) Preterm formula. Figures represent the percentage of weight gain as protein or fat.

All the components of the energy equation have been either measured or estimated in preterm infants, in order to calculate the optimal energy intake. The daily energy cost of activity is around 10 kcal/kg; the cost of thermogenesis, 10 kcal/kg/24 h (depending on thermal environment); and, at growth rates of 15 g/kg/24 h, the cost of tissue synthesis is about 10–20 kcal/kg ( ; ). The remaining component of energy expenditure, basal metabolic rate (BMR), cannot be measured in preterm infants as it requires at least a 12-hour fast. However, total energy expenditure, which includes all four of these components, has been measured over 5 days with both the doubly labelled water method and indirect calorimetry in four preterm infants ( ) growing at a mean rate of 15 g/kg/24 h, and was 58 kcal/kg/24 h (range 57–60). Other studies making measurements over 24-hour periods have produced similar figures, in the range 50–70 kcal/kg/24 h. If these figures are entered into equation (2):

a value for BMR of between 17 and 40 g/kg/24 h is obtained ( ).

In order to estimate energy requirements, it is necessary to consider both the energy stored and energy excreted. In the study by , energy stored was 59 kcal/kg/24 h (which is about twice the figure for in utero energy deposition and substantially higher than ideal estimates for preterm infants, owing to the increased fat deposition seen postnatally). Energy losses in the stools have been estimated to be around 10–30 kcal/kg bodyweight, but massive losses may occur (well in excess of 50% of intake) in sick babies with an immature gut. If these figures are entered into equation (1):

Stay updated, free articles. Join our Telegram channel

Join