After completing this article, readers should be able to: One of the traditional goals of neonatal nutrition is to maintain growth within standardized limits. In the preterm infant, growth between the 10th and 90th percentiles of intrauterine rate has been an ideal goal. More recently, maintenance of lean body mass and bone density; prevention of complications such as chronic lung disease, necrotizing enterocolitis, and infection; and optimization of neurodevelopment and adult health through early nutritional programming have become recognized as more meaningful goals than mere somatic growth.To meet these goals, current nutritional practices require thorough scrutiny in terms of quantity and quality (composition) and how they are customized, if not for the individual patient (which would be ideal), at least for subgroups such as those who are small or appropriate for gestational age, sick or healthy, preterm or term, or male or female.Fetal nutrition and development frequently are used as a template for the low-birthweight infant. A large body of literature on the consequences of fetal undernutrition shows that infants born small for gestational age (SGA) not only have short-term morbidity, but long-term morbidity manifested in what has been termed the “metabolic syndrome.” This syndrome includes abdominal obesity, arterial hypertension, and insulin resistance. Other clinical manifestations can be observed, including thyroid dysfunction, hirsutism, ovarian hyperandrogenism and infertility, dyslipidemia (increased triglyceride concentrations, decreased high-density lipoprotein cholesterol), premature atherosclerosis, hyperuricemia, increased cancer risk, and increased incidence of Alzheimer disease.Metabolic syndrome rapidly is becoming a leading public health problem worldwide. The hypothesis garnered primarily from epidemiologic studies that nutritional aberration in fetal life can “program” the individual for health problems in later life as well as for subsequent generations has prompted questions about long-term and even transgenerational effects of postnatal nutrition. It has been suggested that feeding commercial formulas rather than human milk may predispose the infant to development of metabolic syndrome in adulthood. If this effect does occur, can it be propagated to subsequent generations? Are there certain compositional differences in protein or carbohydrate that predispose to metabolic syndrome when provided in high quantities? Similar to growth-restricted fetuses, many preterm neonates currently suffer significant delays in growth in the first several weeks after birth. Attempts to compensate for these delays are made by using specialized parenteral and enteral formulations that contain higher contents of certain nutrients (eg, protein) than human milk or nonspecialized preterm formulas. Epidemiologic studies and studies in animals are beginning to provide data that raise significant concern that manipulation of early postnatal nutrition designed for “catch-up” growth may lead to metabolic syndrome in adulthood and that metabolic aberrations can propagate to subsequent generations.This very broad area has major public health implications. We focus in this article on current nutritional practices that might affect outcomes in the preterm infant and provide speculations about SGA and appropriate-for-gestational age (AGA) infants. We present a few select animal studies that aid in understanding the mechanisms of how early postnatal under- and overnutrition might affect adult health and be propagated to subsequent generations. We also attempt to provide scientifically based approaches to administration of optimal early postnatal nutrition and point to areas where future investigation is needed.Attempts are being undertaken to make nutritional intake in VLBW infants more physiologic by modeling their intake to what such infants would be receiving in utero. Although this has been accomplished partially for parenteral nutrition, changes still must be made for enteral nutrition. In utero, fetuses receive continuous intravenous nutrition by placental flux. Parenteral amino acid and lipid solutions attempt to mimic the nutritional composition of this placental flux. The fetus also receives about 450 mL/d of amniotic fluid through the gastrointestinal tract during the last trimester of pregnancy. After preterm birth, enteral intakes are delayed because of digestive tract immaturity and the fear of developing necrotizing enterocolitis. Common practice for sick preterm infants involves waiting a few days before instituting intravenous amino acid or lipid intake and progressing slowly at a rate of 0.5 g/kg per day, with frequent starts and stops due to high serum urea nitrogen concentrations, hyperbilirubinemia, “rule out sepsis” episodes, thrombocytopenia, and other questionable practices that are based on either sparse or no scientific basis. These practices frequently result in the infant receiving significantly less protein and lipid compared with the amount that would have been received in utero and is at least partially responsible for the growth retardation evidenced by the shift to the right in the growth curve documented in the first several postnatal weeks in these infants (Fig. 1).Frequently, nutrients are withheld at the times (usually the first 2 to 3 wk after birth) when the infant is most critically ill and nutritional requirements are highest, not only for growth, but for meeting the metabolic demands of critical illness. If and when the infant survives this initial critical period, nutritional intake often is accelerated to attain “catch-up” growth. For example, specialized preterm formulas have been developed that contain higher compositions of various nutrients such as protein, nonlactose carbohydrates, and minerals such as calcium and phosphorus.It is possible that at least some of this early postnatal growth delay can be prevented. A recent trend has been toward providing earlier aggressive nutrition via both the enteral and parenteral routes to prevent postnatal growth delays. This is partially based on our knowledge of the flux of nutrients flowing from the maternal-placental circulation to the fetus and the desire to maintain similar intake in the preterm infant. Current recommendations for parenteral nutrition include a provision of glucose up to 10 to 12 mg/kg per minute, 2 to 3 g/kg per day of intravenous lipid emulsion, and 3 to 4 g/kg per day of crystalline amino acids shortly after birth.A lack of food in the gastrointestinal tract resulting in various complications such as villus atrophy, lack of hormonal stimulation, increased proinflammatory mediator release, and the need for prolonged parenteral nutrition with its attendant complications (eg, sepsis and liver pathology) is becoming increasingly recognized. Accordingly, when VLBW infants reach clinical stability, one of the major goals is to increase the enteral nutrition rapidly to attain catch-up growth similar to the fetal growth rate. Because human milk may not be adequate to reach this goal, it is enriched with fortifiers containing additional carbohydrates (frequently not lactose), lipids, proteins, and micronutrients. Preterm formulas are similarly enriched with more carbohydrates, proteins, and lipids than formulas used for term infants. This higher nutrient density could have both positive and negative consequences.For term neonates, breastfeeding appears to provide health advantages in terms of decreased infections and prevention of metabolic disease such as type 1 diabetes as well as atopy. However, the ideal duration of exclusive breastfeeding is still not known.When term neonates are not breastfed, they receive a term formula. Although there are several similarities in the composition of term formula and human milk, term formula frequently contains more protein. The long-term metabolic consequences of this relatively high protein concentration in formula have not been evaluated.Furthermore, soy formulas that are rich in isoflavone, a natural estrogen, are used commonly. Long-term effects of high estrogen consumption in early life also are not known.For neonates who experience IUGR, the nutritional goal frequently is to attain a rapid catch-up growth, at least before 1 year of age. To accomplish this, enriched human milk and high-caloric and -protein density formulas are provided. This relative calorie and protein overfeeding could have more long-term consequences on infants who experience IUGR than on preterm AGA neonates because of the earlier programming toward “thriftiness” in utero.Many infants who have severe IUGR do not catch up in terms of adult growth. Moreover, follow-up of preterm newborns at 18 to 22 months corrected age shows that 40% still have weights, lengths, and head circumferences less than the 10th percentile. For these children, growth hormone treatment for a few years has been proposed when their height remains below the -3 standard deviations line at 3 years of age. Growth failure is associated with an increased risk of poor neurodevelopmental outcome. Brain volumes of adult survivors of very low birthweight recently have been evaluated in a case-control comparison with sibling controls. The results showed that adults whose birthweights were very low had significantly increased ventricular volume and decreased posterior corpus callosum volume compared with their normal-birthweight siblings.Numerous studies have reported the beneficial effects of supplementing human milk with fortifiers and the use of specialized preterm formulas versus term formulas on neurodevelopment and growth. However, recently Morley and associates showed that infants who experienced IUGR (born between 1993 and 1995) and were fed enriched formula versus term formula had a disadvantage at 9 and 18 months, with a sex interaction at 9 months. Girls fed the enriched formula had significantly lower scores in all subscales and in overall developmental quotient.In addition, studies showing detrimental effects of attempts to attain catch-up growth are emerging. It is not unusual to see infants who were critically ill and undernourished in a neonatal intensive care unit become grossly obese after they are provided nutrient-rich intakes. This is one component of the spectrum of overcompensation. However, the effects may be subtler and may take many years to emerge. Recent studies evaluating long-term effects of early nutrition suggest that there may be long-term detrimental consequences of early aggressive nutrition, especially after attempts to attain catch-up growth. Singhal and colleagues measured fasting concentrations of 32–33 split proinsulin, a marker of insulin resistance, in adolescents born preterm who had participated in randomized intervention trials of neonatal nutrition and in adolescents born at term. Fasting 32–33 split proinsulin concentrations were greater in children given a nutrient-enriched neonatal diet than in those given the lower-nutrient diet. Greatest fasting 32–33 split proinsulin concentrations were associated with greater weight gains the first 2 postnatal weeks. These results suggest that relative overnutrition early in life for children born preterm might have long-term adverse effects on insulin resistance. Ericksson and associates also have shown in the Helsinki cohort (8,760 infants born from 1934 to 1944) that rapid gain in body mass index after 2 years of age was associated with an increased risk of later disease, and this effect was greatest among adults who had slow growth in length between birth and 3 months of age.The relative benefits versus risks of catch-up growth are leading to controversy. Studies in animals should help us understand the mechanisms of the development of metabolic syndrome and subsequently guide us to more rational nutritional regimens for neonates.Because it is difficult to interpret the results in studies of human cohorts, animal studies are important to clarify the mechanisms.Pigs and baboons can be used as animal models for the study of postnatal nutrition, but their use can be expensive and difficult. The litter manipulation model in rodents is used more commonly to study the long-term effects of postnatal under- and overnutrition. In this model, pups are redistributed shortly after birth to either small or large litters, resulting in overnutrition or undernutrition during the suckling period, respectively. McCance showed that rats undernourished during the suckling period followed a permanently diminished growth trajectory, but those undernourished after weaning compensated for the period of slowed growth and resumed the growth trajectory of their normally nourished littermates (Fig. 2).However, the litter alteration technique has significant shortcomings and may not correspond directly to the effects seen in humans. In humans, a permanent diminished growth trajectory is common among infants experiencing severe IUGR. Otherwise, term babies who have a normal birthweight but are undernourished during a few weeks or months appear to catch up.One problem with the techniques used by McCance is that exact volume and nutrient composition received by the pups are difficult to control and measure.The artificial rearing technique, as described by Hall and colleagues, usually called the “pup in the cup” model (Fig. 3), partially circumvents this problem. This model involves placement of a soft catheter into the stomach of the infant rat (usually beginning 4 to 8 d after birth) using an esophageal approach and feeding the pups with a modified rat milk substitute formula, the components of which can be altered. Using this technique, feeding a high-carbohydrate diet throughout the suckling period compared with a composition similar to that of rat milk resulted in several metabolic alterations (Fig. 4). The alterations included increased insulin concentration, increased expression of the gene controlling insulin release (PDX1), and increased hexokinase activity, demonstrating a short-term metabolic response to increased carbohydrate intake.However, despite provision of a similar chow diet by gastrostomy from day 4 to day 20, the metabolic abnormalities seen at 12 days of age, a time during which the high-carbohydrate diet was being provided, persisted at 100 days of age (Fig. 5). In addition to signs of glucose intolerance, animals that were fed the high-carbohydrate diets were obese at 100 days after birth (first observed at day 45). At this time, high-carbohydrate-fed rats weighed approximately 140 g more than their mother-fed counterparts. The high-carbohydrate diet induced an upregulation of several genes involved in glucose metabolism, including insulin, PI3K, and GLUT-2 transporter gene.Females that were fed the carbohydrate-enriched diet during infancy had progeny that continued to present the same metabolic abnormalities (insulin resistance and obesity) as adults when fed the regular diet since birth (Fig. 6). From these studies, it can be inferred that long-term effects that can span from one generation to the next occur with postnatal perturbations in nutrition. The normal development of a multicellular organism is driven by genetic instructions acquired at conception. Yet, during the early critical period of life, the body also has the ability to respond to environmental situations. Thus, our genetic constitution in terms of nucleotide base pairing appears to be only a part of what constitutes our phenotype. The equation that follows depicts the fact that modern nutritional science is beginning to support Lamarckian concepts that one’s phenotype is shaped not only by genetics, but by environment and interaction of one’s genes with the environment. \batchmode \documentclass[fleqn,10pt,legalpaper]{article} \usepackage{amssymb} \usepackage{amsfonts} \usepackage{amsmath} \pagestyle{empty} \begin{document} \[\mathrm{Phenotype}{=}\mathrm{genetics}{+}\mathrm{environment}{+}{\{}\mathrm{adult}{\}}\ \mathrm{genetics}{\times}\mathrm{environment}{\{}\mathrm{gamete}{\}}\] \end{document} Within this framework, early adaptation to a nutritional stress may change the physiology and metabolism of the organism and its progeny permanently. One of the adaptations evident in animal models of early nutritional stress is altered apoptotic homeostasis. Apoptosis plays a central role in perinatal organogenesis and growth in several organs, including the kidney and brain. Gene expression of the apoptosis-related molecules Bcl-2 and Bax responds to multiple nutrient-related stimuli, such as paracrine growth factors. Bcl-2 is an antiapoptosis protein that attenuates the effects of cytochrome c release from the mitochondria and counters the effects of the proapoptosis protein Bax. Bcl-2 and Bax contribute to the signaling pathways that activate caspase-3, which is necessary for the chromatin condensation and DNA fragmentation that characterize apoptosis.IUGR in the rat affects expression of these proteins in a tissue-specific manner, which leads to long-term effects. For example, altered expression of both Bcl-2 and Bax characterize the IUGR kidney and increase renal apoptosis, which permanently decreases glomeruli number. Similarly, decreased cerebral expression of Bcl-2 and increased vulnerability to perinatal hypoxia permanently decrease mitochondrial number in the IUGR rat brain. Although data from a rat model should be applied to human pathophysiology cautiously, it is intriguing that humans who experienced IUGR are predisposed to suffer from decreased glomeruli and adult-onset hypertension as well as long-term neurodevelopment delay.A more subtle adaptation through which the newborn may respond to altered nutrient delivery involves epigenetics. Epigenetics is related to changes in the three-dimensional structure of chromatin without affecting DNA sequence. These changes are heritable and relatively persistent. Two key determinants of chromatin structure are DNA methylation and covalent modification of histone tails. DNA methylation occurs on the C5 position of cysteines within CpGs, which often are found in promoters (Fig. 7). The presence of DNA methyltransferases allows for propagation of the methylation pattern after DNA replication. Histone tails are modified by acetylation, methylation, phosphorylation, and ubiquitination at specific sites (Fig. 8).The essentially limitless combination of histone site-specific modification has led several authors to speculate on the existence of a “histone code” that contains detailed heritable genetic information. Specific combinations of DNA methylation and histone tail modifications lead to the formation of euchromatin or heterochromatin. The three-dimensional structure of euchromatin allows transcription factor complex access to DNA, thereby permitting subsequent transcription of select genes.Multiple nutrition-sensitive signaling pathways communicate conditions in the extracellular milieu to the nucleus and potentially alter these determinants of chromatin structure. As a result, an early significant nutritional stress can affect DNA methylation and histone covalent modification, which subsequently reprograms the cell’s response to further extracellular signals. For example, it has been demonstrated recently that uteroplacental insufficiency in the IUGR rat increases hepatic S-adenosyl homocysteine (SAH), a key substrate involved in one-carbon metabolism whose levels rise in response to folate deficiency. Interestingly, hepatic SAH levels also correlate inversely with DNA methylation, and the hepatic DNA hypomethylation and histone H3 hyperacetylation characterize the newborn IUGR rat. Furthermore, this DNA hypomethylation and histone H3 hyperacetylation persist beyond the newborn period in association with altered mRNA levels of several important metabolic proteins.Although the findings to date are specific to a model of IUGR, the paradigm of an aberrant intrauterine milieu affecting chromatin structure, thereby altering the functional genome’s response to later environments, is potentially applicable to situations such as gestational diabetes or postnatal nutrient stresses such as an early high carbohydrate load.In summary, among the several molecular mechanisms likely to play roles in determining the effects of early nutritional stresses on adult metabolism, both apoptosis and epigenetic phenomena have been identified through animal models as potentially relevant candidates. An early nutritional stress may cause an immediate loss of a specific subset of cells that permanently affects the newborn or may initiate a reprogramming that unfolds over the lifetime of the infant. Unfortunately, the specific mechanism through which certain nutrients trigger these pathways and how they affect adult phenotype are unknown.For healthy term infants, there appear to be no advantages of formula over human milk. The higher protein concentrations found in commercial formulas have been associated with metabolic syndrome in adulthood. Preliminary studies suggest that breastfed infants have a lower incidence of adult metabolic syndrome. Substitution of lactose with other carbohydrates that may cause a higher glycemic index also may program the metabolism of the infant adversely by mechanisms similar to those described for gastrostomy-fed rat pups receiving high-carbohydrate diets. However, experimental evidence for this speculation is lacking.In sick VLBW preterm infants, limitation of early nutrition not only may lead to devastating short-term consequences such as increased susceptibility to infection, lack of organ growth (eg, brain) during a critical period, and pathologic bone fractures, but also to long-term consequences such as poor neurodevelopment. VLBW females catch up in growth by 20 years of age; VLBW males remain significantly shorter and lighter than controls. Because catch-up growth may be associated with metabolic and cardiovascular risk later in life, these findings may have implications for the future adult health of VLBW survivors in whom aggressive attempts are made to compensate for poor initial growth. Even though some evidence suggests that higher intakes are associated with evidence of type 2 diabetes in early adulthood, the complications of early undernutrition, such as poor neurodevelopment, appear to outweigh these metabolic problems. Hence, early aggressive nutrition should remain a mainstay in therapy for VLBW infants.One key aspect is to prevent extrauterine growth retardation with early aggressive nutrition to attain growth similar to that which would be attained in utero and prevent the initial growth delay. This lessens the need for extremely high-density caloric and protein supplementation to attain catch-up growth. As the VLBW preterm infant matures, a common error is to assume that the nutritional requirements on a per-weight basis stay the same as when the infant was in the early stages of postnatal life. The requirements decrease as the infants get older, and intake should be decreased on a per-weight basis.There is now strong evidence of both short- and long-term adverse effects of fetal malnutrition. Postnatal malnutrition also has long-term effects on health, including metabolism, neurologic development, and the cardiovascular system. Catch-up growth in certain circumstances may be a potential cause of metabolic syndrome. Further studies are required to assess the hypothesis that a “thrifty” phenotype/genotype may result from postnatal nutrition.Numerous questions are being raised. Is a higher level of obesity seen in children and adults who were formula-fed during infancy? If so, what is the mechanism? Several studies have demonstrated differences in susceptibility to growth aberrations between boys and girls who were fed similarly as infants. Why are there differences in susceptibility to metabolic syndrome between boys and girls? Should we be feeding infant boys differently than infant girls? How do we achieve full enteral nutrition safely in VLBW infants shortly after birth? Is there a specific duration of the critical window during which the nutrition must be adjusted very carefully? Is the critical window the same for preterm as for term infants? Does under- or overfeeding one specific nutrient have more impact than others? Hopefully, answers will be found to these important questions in the near future.