Multicentre trial on feeding low birthweight infants: effects of diet on early growth

Taurine in pediatric nutrition

The past 20 years have seen the status of taurine change from an end product of methionine and cysteine metabolism and substance conjugated to bile acids to that of an important, and sometimes essential, nutrient. It is now added to most synthetic human infant formulas and pediatric parenteral solutions throughout the world. This article describes the research that led to this end.

Maternal smoking and blood pressure in 7.5 to 8 year old offspring

Reduced fetal growth in babies born preterm may be associated with reduced later blood pressure, but in children born at term, higher blood pressure. It was hypothesised, therefore, that maternal smoking in pregnancy, associated with reduced fetal growth, programmes later blood pressure differentially according to length of gestation. Six hundred and eighteen children born preterm and now aged 7.5 to 8 years were studied prospectively. Systolic blood pressure in children from smoking compared with non-smoking mothers was significantly lower in those born before 33 weeks' gestation and significantly higher in those born at 33 or more weeks. Within the range 0-40 cigarettes per day until delivery (after adjusting for potentially confounding factors, including social class and current weight) each 10 was associated with a 1.5 mm Hg fall and 2.9 mm Hg rise in pressure for children born below or above 33 weeks' gestation respectively. Similar though smaller differences were seen in diastolic pressure. These data support our hypothesis that later effects of insults impairing fetal growth are gestation dependent, and provide the first evidence that maternal smoking may have long term consequences for blood pressure in children.

Influence of early diet on outcome in preterm infants

Despite intensive research in infant nutrition over the past 50 years, uncertainty exists in nearly every major area of practice. A key factor in this uncertainty has been the lack of knowledge on whether diet or nutritional status in early life has a long-term or permanent influence on health, growth or performance. The possibility that early nutrition has long-term consequences in man has been much debated. There have been limited opportunities to perform formal randomized studies on the effect of early nutrition in humans and many studies have been flawed by problems with study design. Infants born preterm are a special group. At the start of our study in 1982, evidence on which to base choice of diet was inconsistent and related only to short-term outcome, and diets available for such babies differed greatly in nutrient content. In this group it was both ethical and practical to conduct a formal, randomized trial of early diet and outcome and the results were clearly needed for management decisions. We have undertaken a long-term prospective outcome study on 926 preterm infants randomly assigned to the diet received in the neonatal period. Surviving children have been followed at 9 months, 18 months and now 7.5-8 years of age. Our findings suggest that human milk may contain factors which promote brain growth or development and also bone mineralization later in childhood. Outcome data from the randomized trials show that a very brief period of dietary manipulation (on average for the first 4 weeks of life) influences later development.

Feeding the preterm infant

The sick preterm infant has special nutritional needs, and the provision of appropriate nutrition is now recognised as an important part of the increasingly intensive management of this population. "Optimal" nutrition is difficult to define for an individual infant, but prospective randomised studies have shown that the early diet given to preterm infants can have a major impact on their neurological development and growth, with the best outcomes for those receiving either a preterm formula or maternal milk fortified with a range of nutrients. In contrast, diets suitable for term infants do not meet the needs of small preterm infants, either in the short or longer term, and should not be used. This article reviews the nutritional needs of the preterm infant and outlines the strategies commonly used in this area of neonatal intensive care.

Does early nutrition in infants born before term programme later blood pressure?

OBJECTIVES

To test whether nutrition early in infants' development programmes later blood pressure and whether the reported relation between low birth weight and later high blood pressure is due to poor nutrition or growth before full term.

DESIGN

Prospective randomisation of preterm infants to early diets differing greatly in nutrient content in four parallel multicentre trials, with blinded follow up 7.5-8 years later.

SETTING

Neonatal units at Cambridge, Ipswich, King's Lynn, Norwich, and Sheffield.

SUBJECTS

758 children weighing under 1850 g at birth.

MAIN OUTCOME MEASURE

Blood pressure at age of 7.5-8 years.

RESULTS

There were major differences in nutrient intake from randomised diets (preterm formula v standard formula and preterm formula v donor breast milk; in each case with or without mother's milk), but follow up showed no differences in later blood pressure. Individual subjects showed large variation in protein and energy intakes and in growth performance, including degrees of growth failure seldom seen in utero, but these factors were also unrelated to later blood pressure.

CONCLUSION

Extremes of nutritional intake and growth performance in preterm infants do not programme later blood pressure at 7.5-8 years of age. These findings do not support the hypothesis that high blood pressure has early nutritional origins. We suggest that the long term rise in blood pressure reported in individuals who had low birthweight (at full term) is not, as previously speculated, due to poor fetal nutrition or growth as such.

Size at birth and later blood pressure

Adults who were small at birth have been reported to have higher systolic blood pressure and risk of ischaemic heart disease in later life. Conversely, in 616 subjects of less than 34 weeks' gestation and 1850 g birth weight, systolic and diastolic blood pressures at 7.5 to 8 years of age decreased with decreasing weight for gestational age and were not related to birth weight. These data do not support the view that fetal growth retardation before 34 weeks programmes for increased blood pressure in later life.

Long-term associations with infant feeding in a clinically advantaged population of babies

Information on infant feeding and child development was obtained from the 1970 British Births Survey and subsequent follow-ups at five and 10 years of age. A clinically advantaged subsample of infants who were either exclusively bottle-fed or exclusively breast-fed for at least three months were selected in order to minimise biases against a disadvantaged bottle-fed group. Only in the area of educational ability and attainment were differences observed as a function of the child's feeding patterns as an infant. Higher vocabulary test scores at five years and higher British Ability Scales test scores at 10 years were significantly associated with exclusive breast-feeding for more than three months. These data support and extend results obtained in previous studies on long-term developmental differences between breast-fed and bottle-fed children.

A randomised multicentre study of human milk versus formula and later development in preterm infants

Whether breast milk influences later neurodevelopment has been explored in non-randomised studies, potentially confounded by social and demographic differences between feed groups. Here in a strictly randomised prospective multicentre trial, Bayley psychomotor and mental development indices (PDI and MDI) were assessed at 18 months postterm in survivors of 502 preterm infants assigned to receive, during their early weeks, mature donor breast milk or a preterm formula. These diets were compared as sole enteral feeds or as supplements to the mother's expressed breast milk. No differences in outcome at 18 months were seen between the two diet groups despite the low nutrient content of donor milk in relation to the preterm formula and to the estimated needs of preterm infants. These results contrast with those reported from our parallel two centre study that compared infants randomly assigned a standard term formula or the preterm formula during their early weeks; those fed standard formula, now regarded as nutritionally insufficient for preterm infants, were substantially disadvantaged in PDI and MDI at 18 months post-term. It is shown here that infants from that study fed solely on standard formula had significantly lower developmental scores at 18 months than those fed on donor breast milk in the present study; yet the standard formula had a higher nutrient content than the donor milk. Thus, donor milk feeding was associated with advantages for later development that may have offset any potentially deleterious effects of its low nutrient content for preterm infants. As these outcome advantages were not confounded by the social and educational biases usually associated with mothers' choice to breast feed, our data add significant support to the view that breast milk promotes neurodevelopment.

The influence of sodium on growth in infancy

Sodium (Na) is an important growth factor, stimulating cell proliferation and protein synthesis and increasing cell mass. Sodium chloride (NaCl) deprivation inhibits growth, as reflected by reduced body and brain weight, length, muscle and brain protein and RNA content and brain lipid content compared with controls. This is not due to deficiency of other nutrients since control and experimental diets were identical except for NaCl content. Subsequent NaCl supplementation restores growth velocity to control values but does not induce "catch-up" growth. In humans, salt loss causes growth failure and subsequent salt repletion improves growth. Preterm infants < 32 weeks' gestation at birth are renal salt losers in the first 2 weeks of post-natal life and are vulnerable to hyponatraemia. This can be prevented by increasing Na intake, which also produces accelerated weight gain that persists beyond the period of supplementation. Early nutrition in preterm infants can affect subsequent growth and also cognitive function: this is probably multifactorial, but NaCl intake differed substantially between study groups and is likely to be an important factor. The mechanism whereby Na promotes cell growth is not understood, but stimulation of the membrane Na+,H(+)-antiporter with alkalinization of the cell interior is a likely possibility.

Neuropsychiatric development: two case reports about the use of dietary fish oils and/or choline supplementation in children

Choline supplementation has been used with moderate success in subgroups of adult patients with neuropsychiatric and medical problems. The dietary fish oils have also been used in adults with hypercholesterolemia. We report on two young children with multiple neurodevelopmental delays, one who responded to choline and eicosapentaenoic acid, and the other to choline alone. A brief discussion about choline's metabolic pathways and benefits is included.

Hyperphenylalaninaemia and outcome in intravenously fed preterm neonates

Hyperphenylalaninaemia is likely to have occurred in many infants fed the intravenous amino acid solution Vamin 9. In this study of 336 preterm infants plasma phenylalanine was measured weekly during their hospital stay. Reference data on plasma phenylalanine were prepared for 243 infants who did not receive Vamin. Only 1% of these infants had a peak plasma phenylalanine concentration greater than 150 mumol/l (maximum 202 mumol/l) compared with 23% in 93 infants fed Vamin 9, seven of whom had concentrations > 300 mumol/l (maximum 704 mumol/l). High concentrations only occurred when the total energy to protein energy ratio in the intravenous solutions decreased to less than 8.5:1 and always occurred with a ratio less than 6.5:1, implying that hyperphenylalaninaemia may be minimised with an intravenous energy intake of greater than 34 kcal (142 kJ)/g protein. Nevertheless, follow up at 18 months post-term showed that increased plasma phenylalanine in this instance was not associated with any impairment of the Bayley mental development index (or subscales including fine motor, cognitive, or language development), the psychomotor development index, or the social maturity quotient. Thus, despite theoretical concern, an adverse outcome after hyperphenylalaninaemia induced by intravenous feeding has not been observed.

Development of bowel habit in preterm infants

The bowel habits in 844 preterm infants were observed during the first 56 days after birth. Delay in the passage of meconium beyond the second day occurred in 32%, and there was an inverse relation between gestational age and the day of first bowel action. Thereafter the increase in frequency of stools passed each day was related directly to the volume of milk ingested. Unfed infants had a modal defecation rate of one stool each day. For each 50 ml/kg increase in the volume of milk ingested the infants showed a further increase of one stool passed each day. Infants fed with human milk passed a greater number and softer stools than those who received cows' milk formula. In the absence of milk feeds an intrinsic pattern of large bowel motor activity, present as early as 25 weeks' gestation, ensures a defecation rate of one stool each day. The effect of enteral nutrition on this inherent motility is governed by the volume and composition of milk feeds, independently of gestational age.

Recent advances in breast feeding

Increasingly, epidemiological studies are showing the benefits of exclusive breast-feeding to infants in industrialized countries, as they have previously for those in developing countries. In this review we have focused on several recent developments, in particular on the interplay between infant behaviour and maternal physiology in regulating breast milk supply. This will illustrate that many past and current clinical problems concerning breast-feeding are likely to be of iatrogenic origin, the products of imposing arbitrary rules for breast-feeding management. In order for the protective benefits to be fully realized it is necessary to ensure that breast-feeding is free from prescriptive practices. We will illustrate the key role played by the infants' control of appetite in the hope of encouraging clinicians to empower infants to exercise more control over the feeding process.

Delayed surgery for congenital diaphragmatic hernia: neurodevelopmental outcome in later childhood

The long term neurodevelopmental outcome was assessed in 23 survivors born with congenital diaphragmatic hernia who had been managed by an elective delay in surgical repair after a period of stabilisation. This cohort was treated in one neonatal surgical unit between 1983 and 1989 by a single team of surgeons and anaesthetists. All children underwent comprehensive neurological, developmental, and anthropometric assessment at a mean age of 56 (range 18-94) months. Two children (9%) had major disability (one with hemiplegia and one with a lower limb monoplegia) and two further children had minor disabilities (one had partial sightedness and squint, the other squint only). The mean developmental quotient (DQ) for the group was 108 (SD 10.8) and none had developmental delay (defined as DQ < 70). Infants who had spent more time in hospital, or had had a longer duration of ventilation, tended to have lower weights and lower occipitofrontal circumference centiles in later childhood. Preoperative stabilisation and delayed surgery for congenital diaphragmatic hernia is not associated with an impaired neurodevelopmental outcome.

Two different low birth weight formulae compared

The performance of two different, commercially available, low birth weight formulae feeds was compared in preterm infants. The aim of the study was to determine the effect of compositional differences on tolerance, stool frequency and consistency, fat balance and weight gain. Inborn infants with birth weight less than 1500 g were randomised at birth to receive Prematil or Osterprem. Thirty infants received more than 900 ml/kg per week of designated formula alone during a total of 70 weeks of study. Three day fat balance was performed on 23 infants. Osterprem contains 40% more fat than Prematil. The composition of this fat is different in that Osterprem contains no medium chain triglycerides (MCT) compared to 30% in Prematil. Clinical evaluation demonstrated that Osterprem is associated with a significantly higher mean energy intake compared to Prematil (3442 and 3127 kJ/kg per week) but mean weight gain is not significantly different (123 and 112 g/kg per week). Mean stool frequency is higher on Osterprem (20.5 and 14.5 stool/week) and the consistency of stools firmer. This is attributable to a higher mean fat output (2.3 and 0.9 g/kg per day) secondary to the higher fat content of the feed and lower mean absorption (71.6 and 83.5%). Both feeds are well tolerated. The study also confirms that absorption of unsaturated fatty acids is inversely proportional to chain length.

Randomised trial of nutrition for preterm infants after discharge

In a randomised double blind trial, the effect on growth and clinical status of a nutrient enriched 'post-discharge' milk formula versus a standard term formula, was compared in 32 exclusively bottle fed preterm infants. The formulas were used as the sole milk intake up to a postnatal age of 9 months. Significant increases in linear growth and weight gain were observed in the infants who received the enriched diet. There were no differences in vomiting, posseting, or bowel habit between the groups. Formula volumes ingested were similar between diet groups, indicating that the difference in formula composition did not affect the infants' regulation of intake. These preliminary data suggest that there is a role for specially designed formulas for preterm infants after discharge from hospital.

Breast milk and subsequent intelligence quotient in children born preterm

There is considerable controversy over whether nutrition in early life has a long-term influence on neurodevelopment. We have shown previously that, in preterm infants, mother's choice to provide breast milk was associated with higher developmental scores at 18 months. We now report data on intelligence quotient (IQ) in the same children seen at 7 1/2-8 years. IQ was assessed in 300 children with an abbreviated version of the Weschler Intelligence Scale for Children (revised Anglicised). Children who had consumed mother's milk in the early weeks of life had a significantly higher IQ at 7 1/2-8 years than did those who received no maternal milk. An 8.3 point advantage (over half a standard deviation) in IQ remained even after adjustment for differences between groups in mother's education and social class (p less than 0.0001). This advantage was associated with being fed mother's milk by tube rather than with the process of breastfeeding. There was a dose-response relation between the proportion of mother's milk in the diet and subsequent IQ. Children whose mothers chose to provide milk but failed to do so had the same IQ as those whose mothers elected not to provide breast milk. Although these results could be explained by differences between groups in parenting skills or genetic potential (even after adjustment for social and educational factors), our data point to a beneficial effect of human milk on neurodevelopment.

The influence of nutrition on neonatal respiratory muscle function

Advances in neonatal intensive care have resulted in increasing survival of very small babies, who often require prolonged periods of mechanical ventilation. These babies have limited nutritional reserves, and are difficult to feed by either parenteral or enteral routes. This review article discusses the interaction between undernutrition and respiratory muscle function in the critically ill preterm baby.

High folate intakes related to zinc status in preterm infants

The former practice of giving 1 mg (2.27 mumoles) oral folic acid daily to premature infants receiving enteral feeds was assessed with respect to zinc status in Cambridge, United Kingdom. A group of 60 preterm infants, 80% of whom were receiving 1 mg oral folic acid daily, were studied for up to the first 16 weeks of life. Plasma folate and plasma zinc were measured for each subject. A significant inverse relationship was found between the maximum attained serum folate level and the minimum attained serum zinc level, (t = 5.0, 58 df, P less than 0.0001). This remained significant after corrections had been made for gestational age at birth, fetal growth retardation, birth weight, sex, diet, assisted ventilation and length of time to full enteral feeding. The hypothesis that very high folate intakes may adversely affect serum zinc levels and, by inference, zinc status in preterm infants could not be rejected. Caution is therefore advised when prescribing such very high folate doses daily for small preterm infants.

Effect of protein/energy ratio on growth and behavior of premature infants: preliminary findings

Premature infants weighing less than 1550 gm at birth were randomly assigned to receive one of three formulas identical in composition except for protein content (2.2, 2.7, and 3.2 gm.100 kcal-1) to determine the effects on growth, protein nutritional status, and behavior. Data collected for 2 weeks from the time of achieving an enteral energy intake of 100 kcal.kg-1.day-1 included measurements of weight, length, head circumference, and skin-fold thickness, and concentrations of plasma amino acids, serum total protein, prealbumin, retinol-binding protein, and urea nitrogen. In a subset of infants, behavior was assessed at the end of the feeding study with the Neonatal Behavior Assessment Scale. Except for the concentrations of plasma amino acids, there were no significant differences in growth or in other biochemical measurements among the three groups, but there were significant differences in the orientation, habituation, and stability clusters of the behavior assessment. Further, there were significant correlations between the plasma amino acid values and the behavioral clusters. These preliminary data suggest a relationship between protein intake in the neonatal period and behavioral outcome at the end of the feeding period in the absence of differences in growth and gross markers of protein nutritional status. The behavioral items noted to differ among the groups may indicate later cognitive outcome; detailed studies about behavioral responses to neonatal dietary intakes and later outcome seem indicated.

Malnutrition in the premature infant

During the first few days of life, the ill premature infant is usually subjected to acute semistarvation because the provision of nutritional support is considered cumbersome and unnecessary. However, the absence of readily recognizable adverse effects of semistarvation does not rule out the existence of significant short-term adverse effects, nor does it rule out possible adverse sequelae in the long run. Similar concerns pertain to the later neonatal period, during which nutritional deprivation is less severe but of longer duration. Evidence is presented that qualitative malnutrition, characterized by inadequate intake of protein and relatively excessive intake of energy, is common with current feeding regimens and is responsible for increased body fat deposition in growing small premature infants.

Salt and the newborn kidney

Renal function differs in term infants from that in adults, with lower glomerular filtration rate (GFR) and reduced proximal tubular reabsorption of sodium (Na) and water: nevertheless, it is adequate for their needs. This is not true of very preterm infants in whom hyponatraemia is common. Animal studies have shown that Na+, K(+)-ATPase and the Na+/K+ exchanger are poorly expressed at birth with rapid postnatal rises. Cell receptors for hormones that influence tubular Na transport are less numerous in the premature infant than later in life: intracellular second messenger systems may also be immature. The low GFR is due to vasoconstriction and may be necessary to prevent water and electrolyte wasting due to tubular overload. The hyponatraemia of prematurity could, in principle, be due either to Na loss or water excess and can be prevented either by giving additional Na or by restricting water intake. Na supplementation causes relative volume expansion (VE), water restriction volume contraction (VC); this is demonstrated by the effect of the two approaches on weight gain and on the levels of vasoactive hormones in the blood. We argue that moderate VE is more physiological than VC, both in attempting to simulate intrauterine conditions and in consideration of the infant's nutritional needs. The much less common complication of hypernatraemia is usually due to abnormal water loss and should be prevented by increasing water intake appropriately. The above applies to well, preterm babies: sick preterm infants are much more variable in their Na and water requirements than well infants of comparable gestation and weight and each needs an individually tailored regimen based on frequent clinical assessment and laboratory measurement.

Breast milk and neonatal necrotising enterocolitis

In a prospective multicentre study on 926 preterm infants formally assigned to their early diet, necrotising enterocolitis developed in 51 (5.5%). Mortality was 26% in stringently confirmed cases. In exclusively formula-fed babies confirmed disease was 6-10 times more common than in those fed breast milk alone and 3 times more common than in those who received formula plus breast milk. Pasteurised donor milk seemed to be as protective as raw maternal milk. Among babies born at more than 30 weeks' gestation confirmed necrotising enterocolitis was rare in those whose diet included breast milk; it was 20 times more common in those fed formula only. Other risk factors included very low gestational age, respiratory disease, umbilical artery catheterisation, and polycythaemia. In formula-fed but not breast-milk-fed infants, delayed enteral feeding was associated with a lower frequency of necrotising enterocolitis. With the fall in the use of breast milk in British neonatal units, exclusive formula feeding could account for an estimated 500 extra cases of necrotising enterocolitis each year. About 100 of these infants would die.

Plasma prolactin and clinical outcome in preterm infants

Plasma prolactin was measured weekly in 280 preterm infants. The complex gestational age dependent pattern of postnatal prolactin release has been defined and reference standards provided. Plasma prolactin was higher in girls, with increasing divergence between the sexes from the third week onwards, and higher after two weeks, in infants of mothers with pregnancy related hypertension. Diet, assigned randomly, exerted a major effect on plasma prolactin, with significantly higher values in infants fed donor breast milk or standard formula than in those fed a protein, energy, and mineral enriched preterm formula. After adjusting for confounding factors, infants with the lowest plasma prolactin concentrations (less than 1000 mU/l, 32.9 micrograms/l) occurring usually at a nadir between days 5 and 12, showed a 120% increase in the duration of ventilatory assistance required, a 20% increase in the number of days to attain full enteral feeds, and a 30% decrease in length gain. We suggest preterm birth disrupts the normal perinatal pattern of prolactin release and that those infants who develop relatively low plasma concentration have an adverse outcome. Our data add to the broader debate on whether preterm infants require multiple endocrine replacement treatment.

Urinary and serum urea as indicators of protein metabolism in very low birthweight infants fed varying human milk protein intakes

Urea concentrations in serum and urine were measured in 28 growing, very low birth weight, appropriate-for-gestational age infants fed varying human milk protein intakes (range 1.7 to 3.9 g/kg/day). We found a high correlation between serum urea values at the end of the study and mean protein intake (rs = 0.85, p less than 0.001) and between urinary urea concentrations in eight-hour urine collections and protein intake (rs = 0.81, p less than 0.001). All serum and urine urea values were below 1.6 and 18 mmol/l, respectively, at protein intakes less than 3 g/kg/day. Higher protein intakes caused higher serum and urinary urea concentrations. We also found a strong correlation between the individual serum and urinary urea values at the end of the study (rs = 0.90, p less than 0.001). The presented data are consistent with the growth data previously reported and indicate that inadequate or excessive protein intakes can be detected by measurement of urea concentrations in serum and/or urine. If urine urea samples alone can be used for estimating optimal protein intake, painful blood sampling procedures could be obviated.

Early diet in preterm babies and developmental status at 18 months

424 preterm infants were randomly assigned a standard "term" formula or a nutrient-enriched "preterm" formula as sole diets (trial A) or as supplements to mother's own expressed milk (trial B) for a median of 4 weeks postnatally. 18 months post term, blind evaluation of 377 survivors showed that those previously fed preterm rather than term formula had major developmental advantages, more so in motor than mental function; the advantages, in both mental and motor scores, were especially striking in small-for-gestational-age infants and in males. For motor development index in trial A, this advantage was 15 points; in infants born small for gestation, it was 23 points (nearly 1.5 SD). Moderate developmental impairment (developmental index less than 85), notably motor impairment, was considerably more common in the group fed term formula. Infants fed preterm formula also had a small benefit in social maturity quotient. Thus, a short period of early dietary manipulation in preterm infants had major consequences for later development, which suggests that the first weeks may be critical for nutrition.

Food and drug reactions, wheezing, and eczema in preterm infants

Allergic reactions were investigated in 777 preterm infants who were randomly assigned to early diet and followed up to 18 months post term. Wheezing or asthma was common (incidence 23%); it was associated with neonatal ventilation, maternal smoking, and a family history of atopy and was unexpectedly reduced in babies born by caesarean section. Even in non-ventilated infants, the incidence of subsequent wheezing was 18%, rising to an estimated 44% (using logistic regression) when the foregoing risk factors (excluding ventilation) were present. Eczema occurred in 151 infants (19%) and was strongly associated with multiple pregnancy (30% incidence in twins or triplets). Reactions to cows' milk (incidence: 4.4% from detailed history; 0.8% confirmed by challenge), other foods (10%), and drugs (5%) were within the range reported in full term infants. Milk and food reactions were associated with multiple pregnancy (19%) and a family history of atopy. Reactions to drugs were least likely to occur in infants who had been ventilated and were on multiple medications in the neonatal period, suggesting that drug tolerance may have developed. We speculate that preterm infants may be a high risk group for asthma and eczema, which could imply an association between atopy and prematurity.

Early diet of preterm infants and development of allergic or atopic disease: randomised prospective study

OBJECTIVE

To study the effect of early diet on the development of allergic reactions in infants born preterm.

DESIGN

Two randomised prospective trails. In trail A infants were randomly allocated banked donor milk or preterm formula as their sole diet or (separately randomised) as a supplement to their mother's expressed breast milk. In trial B infants were allocated term or preterm formula. A blind follow up examination was done 18 months after the expected date of birth.

SETTING

Neonatal units of hospitals in Cambridge, Ipswich, King's Lynn, Norwich, and Sheffield. Outpatient follow up.

PARTICIPANTS

777 Infants with a birth weight less than 1850 g born during 1982 to 1984.

MAIN OUTCOME MEASURES

Development of eczema, allergic reactions to food or drugs, and asthma or wheezing by nine and 18 months after term. Whenever possible the observations were confirmed by rechallenge or clinical examination.

RESULTS

At 18 months after term there was no difference in the incidence of allergic reactions between dietary groups in either trial. In the subgroup of infants with a family history of atopy, however, those in trial A who received preterm formula rather than human milk had a significantly greater risk of developing one or more allergic reactions (notably eczema) by 18 months (odds ratio 3.6; 95% confidence interval 1.4 to 9.1).

CONCLUSIONS

Feeding neonates on formulas based on cows' milk, including those with a high protein content, did not increase the overall risk of allergy. Nevertheless, in the subgroup with a family history of atopy early exposure to cows' milk increased the risk of a wide range of allergic reactions, especially eczema.

Does early diet program future outcome?

Despite intensive research in infant nutrition throughout this century, uncertainty persists over nearly every major aspect of practice. Reasons for this uncertainty are analysed and the need for prospective studies on the later effects of early diet is justified. Evidence is presented that nutrition at a critical or sensitive period in early life predetermines future metabolism, performance and morbidity in animal models. Similar data in Man have been collected mainly in retrospective and flawed studies and are correspondingly less convincing. A prospective multicentre randomized study on preterm infant feeding is discussed to illustrate the feasibility of conducting long-term outcome studies in Man and to emphasize the critical importance of outcome data as a basis for clinical practice. Preliminary evidence from this study suggests that the way a preterm infant is fed, in just the early weeks post partum, may have a major impact on later growth and development.

Birthweight ratio and outcome in preterm infants

The association between birthweight ratio and outcome was investigated in 429 infants born before 31 weeks' gestation. Birthweight ratio was calculated in each case as birth weight divided by mean birth weight for gestation (from reference data). It was shown that a given ratio corresponded to the same birth centile across the gestational age range studied; a ratio of 0.8 corresponding to the 10th centile. There was a linear relationship between birthweight ratio and requirement for mechanical ventilation and postneonatal mortality. Birthweight ratio was also strongly and linearly related to body weight, length, and head circumference at 18 months' corrected age. Overall, there was no association between this ratio and neurodevelopmental outcome to 18 months. However, the subgroup with the largest weights for gestation (birthweight ratio greater than or equal to 1.1), had significantly higher language subscores than all the other children. Our data show that conventional dichotomous categorisation of preterm infants into small or appropriate for gestation is inadequate when exploring the association between size for gestation and outcome.

Early diet in preterm babies and developmental status in infancy

Few data from randomised prospective studies address whether early diet influences later neurodevelopment in man. As part of a larger multicentre trial, 502 low birthweight infants were assigned randomly, for a median of 30 days, to receive a preterm formula or unfortified donor breast milk as sole diets or as supplements to their mothers' expressed milk. Surviving infants were assessed at nine months after their expected date of delivery without knowledge of their feeding regimen. The mean developmental quotient was 0.25 standard deviations lower in those fed donor breast milk rather than preterm formula. In infants fed their mother's expressed milk, however, the disadvantage of receiving banked milk compared with preterm formula as a supplement, was greater when the supplement was over half the total intake, and approached five points, representing 0.5 standard deviations for developmental quotient. Infants fed donor breast milk were at particular disadvantage following fetal growth retardation, with developmental quotients 5.3 points lower. We suggest that the diet used for low birthweight babies over a brief, but perhaps critical, postnatal period has developmental consequences that persist into infancy; infants who are small for gestational age are especially vulnerable to suboptimal postnatal nutrition.

Human milk for low birthweight infants: immunology, nutrition and newer practical technologies

Human milk has a number of unique properties which can be of benefit to high risk low birthweight and preterm infants. Immunological features are of particular importance. Growth is better with the use of the mother's own milk than with pooled bank milk. The quality of the latter can be improved by various measures noted in the text. Growth can be further improved by enriching the milk with human milk factors, powder formula or protein supplements. As breast-feeding is best for infants, the maternal milk supply should be maintained during the hospital stay of infants by frequent expression, galactogogues used where necessary under supervision, encouraging maternal contact with the infant, promotion of suckling initially on the "empty" breast after expression of milk, and subsequently institution of direct breast-feeding. With proper interventions and motivation of both staff and mothers, even very low birthweight infants below 1,500 gm can go on, in time, to exclusive breast-feeding.

High alkaline phosphatase activity and growth in preterm neonates

In a study on 857 infants born preterm, high peak plasma alkaline phosphatase activity was independently related to slower growth rate in the neonatal period, and to a highly significant reduction in attained length at 9 months and 18 months post term. At 18 months the deficit in body length associated with peak neonatal plasma alkaline phosphase activity of 1200 IU/l or more was 1.6 cm (95% confidence interval 0.9 to 2.3 cm) after adjusting for confounding factors. The strength and magnitude of this association between high plasma alkaline phosphase activity and body length was greater than that for any other factor identified, including the infant's sex and the presence of fetal growth retardation. Data are presented that support the view that the high plasma alkaline phosphatase activity reflected early bone mineral substrate deficiency resulting in metabolic bone disease. We speculate that even silent early bone disease may interfere with the control of subsequent linear growth and emphasise the potential importance of providing preterm infants, especially those fed human milk, with adequate substrate for bone mineralisation.

Growth and development in premature twins

A total of 476 infants (386 singletons and 90 twins) born before 32 weeks' gestation were studied to compare long term growth and development in twins and singletons. At 18 months, after adjusting for confounding social, obstetric, and neonatal factors, twins were not disadvantaged in their neurodevelopmental status, but were 1.6 cm shorter than singletons and had thicker triceps and subscapular skinfolds. No significant differences were found between first and second born twins in later growth or development at 18 months post-term. While preterm twins may have an inherent disadvantage in linear growth it is suggested that in other respects twinning is not a risk factor in preterm infants.

Adverse neurodevelopmental outcome of moderate neonatal hypoglycaemia

There has been considerable debate over whether asymptomatic neonatal hypoglycaemia results in neurological damage. In a detailed multicentre study of 661 preterm infants hypoglycaemia was found to be common. Moderate hypoglycaemia (plasma glucose concentration less than 2.6 mmol/l) occurred in 433 of the infants and in 104 was found on three to 30 separate days. There was considerable variation among the centres, implying differences in decisions to intervene. The number of days on which moderate hypoglycaemia occurred was strongly related to reduced mental and motor development scores at 18 months (corrected age), even after adjustment for a wide range of factors known to influence development. When hypoglycaemia was recorded on five or more separate days adjusted mental and motor developmental scores at 18 months (corrected age) were significantly reduced by 14 and 13 points respectively, and the incidence of neurodevelopmental impairment (cerebral palsy or developmental delay) was increased by a factor of 3.5 (95% confidence interval 1.3 to 9.4). These data suggest that, contrary to general belief, moderate hypoglycaemia may have serious neurodevelopmental consequences, and reappraisal of current management is urgently required.

Mother's choice to provide breast milk and developmental outcome

The association between a mother's choice to provide breast milk and her baby's developmental status at 18 months post term was investigated in 771 low birthweight infants from five centres. Babies whose mothers chose to provide milk had an 8 point advantage in mean Bayley mental developmental index over infants of mothers choosing not to do so. A 4.3 point advantage remained after adjusting for demographic and perinatal factors. A similar finding was derived using a fundamentally different and questionnaire based test (academic scale of Developmental Profile II). Whether this significant residual developmental advantage relates to parental factors or to a beneficial effect of human milk itself on brain development has important implications for the nutritional management of premature babies.

Low plasma triiodothyronine concentrations and outcome in preterm infants

A major association has been found between low plasma triiodothyronine concentrations in preterm neonates and their later developmental outcome. Plasma triiodothyronine concentration was measured longitudinally in 280 preterm infants below 1850 g birth weight. Babies whose lowest recorded concentration was less than 0.3 nmol/l had, at 18 months' corrected age, 8.3 and 7.4 point disadvantages in Bayley mental and motor scales and a 8.6 point disadvantage on the academic scale of Developmental Profile II, even after adjusting for a range of antenatal and neonatal factors known to influence later development. Low concentrations of triiodothyronine were strongly associated with infant mortality, but not after adjusting for the presence of respiratory illness. There was no association between plasma triiodothyronine concentrations and somatic growth up to 18 months, and no association with necrotising enterocolitis or later cerebral palsy. Data on postnatal changes in plasma triiodothyronine concentrations are presented for reference purposes. While cited reference ranges for plasma triiodothyronine concentration appear suitable for well infants above 1500 g birth weight, smaller or ill babies often have very low values for many weeks. Our data are relevant to the debate on endocrine 'replacement' treatment in premature babies.