Nutritional Support for the Pediatric Patient

The Metabolic Response to Stress

The metabolic response to illness due to stressors such as trauma, surgery, or inflammation has been well described, and the magnitude of the response varies according to illness severity. Cuthbertson was the first investigator to realize the primary role that whole-body protein catabolism plays in the systemic response to injury. Based on his work, the metabolic stress response has been conceptually divided into two phases. The initial, brief “ebb phase” is characterized by decreased enzymatic activity, reduced oxygen consumption, low cardiac put, and core temperature that may be subnormal. This is followed by the hypermetabolic “flow phase” characterized by increased cardiac output, oxygen consumption, and glucose production. During this phase, fat and protein mobilization is manifested by increased urinary nitrogen excretion and weight loss. This catabolic phase is mediated by a surge in cytokines and the characteristic endocrine response to trauma or surgery that results in an increased availability of substrates essential for healing and glucose production.

Neonates and children share similar qualitative metabolic responses to illness as adults, albeit with significant quantitative differences. The metabolic stress response is beneficial in the short term, but the consequences of sustained catabolism are significant because the child has limited tissue stores and substantial nutrient requirements for growth. Thus, the prompt institution of nutritional support is a priority in sick neonates and children. The goal of nutrition in this setting is to augment the short-term benefits of the metabolic response to injury while minimizing negative consequences of persistent catabolism. In general, the metabolic stress response is characterized by an increase in net muscle protein degradation and the enhanced movement of free amino acids through the circulation ( Fig. 2.1 ). These amino acids serve as the building blocks for the rapid synthesis of proteins that act as mediators for the inflammatory response and structural components for tissue repair. The remaining amino acids not used in this way are channeled through the liver, where their carbon skeletons are utilized to create glucose through gluconeogenesis. The provision of additional dietary protein may slow the rate of net protein loss, but it does not eliminate the overall negative protein balance associated with injury.

Carbohydrate and lipid turnover are also increased severalfold during the metabolic response. Although these metabolic alterations would be expected to increase overall energy requirements, data show that such an increase is quantitatively variable, modest, and evanescent. Overall, the energy needs of the critically ill or injured child are governed by the severity and persistence of the underlying illness or injury. Accurate assessment of energy requirements in individual patients allows optimal caloric supplementation and avoids the deleterious effects of both underfeeding and overfeeding.

Children with critical illness demonstrate a unique hormonal and cytokine profile. A transient decrease in insulin levels is followed by a persistent elevation, the anabolic effects of which are overcome by increased levels of catabolic hormones (glucagon, cortisol, catecholamines). This overall catabolic state is marked by increases in specific inflammatory cytokines (interleukin [IL]-6, tumor necrosis factor [TNF]-α). Novel ways to manipulate these hormonal and cytokine alterations with an aim to minimize the deleterious consequences induced by the stress response are a focus of research.

Body Composition and Nutrient Reserves

The body composition of the young child contrasts with that of the adult in several ways that significantly affect nutritional requirements. Table 2.1 lists the macronutrient stores of the neonate, child, and adult as a percentage of total body weight. Carbohydrate stores are limited in all age groups and provide only a short-term supply of glucose. Despite this fact, neonates have a high demand for glucose and have shown elevated rates of glucose turnover compared with those of the adult. This is thought to be related to the neonate’s increased ratio of brain-to-body mass because glucose is the primary energy source for the central nervous system. Neonatal glycogen stores are even more limited in the early postpartum period, especially in the preterm infant. Short periods of fasting can predispose the newborn to hypoglycemia. Therefore, when infants are burdened with illness or injury, they must rapidly turn to the breakdown of protein stores to generate glucose through the process of gluconeogenesis. In premature infants, gluconeogenesis is sustained despite provision of parenteral nutrition (PN) with glucose infusion rates higher than endogenous glucose production rate.

Lipid reserves are low in the neonate, gradually increasing with age. Premature infants have the lowest proportion of lipid stores because the majority of polyunsaturated fatty acids accumulate in the third trimester. This renders lipid less available as a potential fuel source in the young child. The most dramatic difference between adult and pediatric patients is in the relative quantity of stored protein. The protein reserve per kilogram of ideal body weight in the adult is nearly twofold that of the neonate. Thus, infants cannot afford to lose significant amounts of protein during the course of a protracted illness or injury. An important feature of the metabolic stress response, unlike in starvation, is that the provision of dietary glucose does not halt gluconeogenesis. Consequently, the catabolism of muscle protein to produce glucose continues unabated. Neonates and children also share much higher baseline energy requirements than adults. In addition, among preterm infants with low birth weight, the birth weight inversely correlates with resting energy expenditure (REE). Clearly, the child’s need for rapid growth and development is a large component of this increase in energy requirement. Moreover, increased heat loss via the relatively large body surface area of the young child and immature thermoregulation in preterm infants further contribute to elevations in energy expenditure.

The basic requirements for protein and energy in the healthy neonate, child, and adult, based on recommendations by the National Academy of Sciences, are listed in Table 2.2 . As illustrated, the recommended protein needs for the infant are two to three times those of the adult. In premature infants, a minimum protein allotment of 2.8 g/kg/day is required to maintain in utero growth rates. The increased metabolic demand and limited nutrient reserves of the infant mandates early nutritional support in times of injury and critical illness to avoid negative nutritional consequences.

An accurate assessment of body composition is necessary for planning nutritional intake, monitoring dynamic changes in the body compartments (such as the loss of lean body mass), and assessing the adequacy of nutritional supportive regimens during critical illness. Ongoing loss of lean body mass is an indicator of inadequate dietary supplementation and may have clinical implications in the hospitalized child. However, current methods of body composition analysis (e.g., anthropometry, weight and biochemical parameters) are either impractical for clinical use or inaccurate in a subgroup of hospitalized children with critical illness. One of the principal problems in critically ill children is the presence of capillary leak, manifesting as edema and large fluid shifts. These make anthropometric measurements invalid, and other bedside techniques have not been adequately validated.

Energy Expenditure During Illness

For children with illness or undergoing operative intervention, knowledge of energy requirements is important for the design of appropriate nutritional strategies. Dietary regimens that both underestimate and overestimate energy needs are associated with negative consequences. Owing to the high degree of individual variability in energy expenditure, particularly in the most critically ill patients, the actual measurement of REE is recommended.

The components of total energy expenditure (TEE) for a child in order of magnitude are REE, energy expended during physical activity (PA), and diet-induced thermogenesis (DIT). The sum of these components determines the energy requirement for an individual. In general, REE rates decline with age from infancy to young adulthood, at which time the rate becomes stable. In children with critical illness, the remaining factors in the determination of total energy requirement are of reduced significance because PA is low and DIT may not be significant.

REE can be measured using direct or indirect methods. The direct calorimetric method measures the heat released by a subject at rest and is based on the principle that all energy is eventually converted to heat. In practice, the patient is placed in a thermally isolated chamber, and the heat dissipated is measured for a given period. This method is the true gold standard for measured energy expenditure. Direct calorimetry is not practical for most hospitalized children, and REE is often estimated using standard equations. Unfortunately, REE estimates using standardized World Health Organization (WHO) predictive equations are unreliable, particularly in critically ill children.

REE estimation is difficult in critically ill or postoperative children. Their energy requirements show individual variation and depend on severity of injury, sedation, and environmental factors. For instance, a mechanically ventilated child with severe traumatic brain injury who is being treated with sedation and neuromuscular blockade would have a much lower energy expenditure than a severely burned child in a nonthermoneutral environment. Infants with congenital diaphragmatic hernia on extracorporeal membrane oxygenation (ECMO) support have been shown to have energy expenditures of approximately 90 kcal/kg/day. Following extubation, the same patients may have energy requirements as high as 125 kcal/kg/day to achieve desired growth velocity at hospital discharge. Although stress factors ranging from 1.0 to 2.7 have been applied to correct for these variations, calculated standardized energy expenditure equations have not been satisfactorily validated in critically ill children. The most recent guidelines for nutrition support in critically ill children, published jointly by the American Society for Parenteral and Enteral Nutrition (ASPEN) and the Society of Critical Care Medicine (SCCM), recommend that if estimating equations are used, the Schofield, WHO, or United Nations University equations may be applied without the addition of stress factors.

Indirect calorimetry measures V O ₂ (the volume of oxygen consumed) and V CO ₂ (the volume of CO ₂ produced) and uses a correlation factor based on urinary nitrogen excretion to calculate the overall rate of energy production. The measurement of energy needs is “indirect” because it does not use direct temperature changes to determine energy needs. Indirect calorimetry provides a measurement of the overall respiratory quotient (RQ), defined as the ratio of CO ₂ produced to O ₂ consumed ( V CO ₂ / V O ₂ ) for a given patient. Oxidation of carbohydrate yields an RQ of 1.0, whereas fatty acid oxidation gives an RQ of 0.7. However, the role of the RQ as a marker of substrate use and an indicator of underfeeding or overfeeding is limited. The body’s ability to metabolize substrate may be impaired during illness, making assumptions about RQ values and substrate oxidation invalid.

Although RQ is not a sensitive marker for adequacy of feeding in individual cases, RQ values greater than 1.0 can be associated with lipogenesis secondary to overfeeding. However, numerous factors, related and unrelated to feeding, can alter the value of a measured RQ in critically ill patients, for example, hyperventilation, acidosis, effects of cardiotonic agents and neuromuscular blocking, and an individual response to a given substrate load, injury, or disease. Furthermore, in the setting of wide diurnal and day-to-day variability of REE in critically ill individuals, the extrapolation of short-term calorimetric REE measurements to 24-hour REE may introduce errors. The use of steady-state measurements may decrease these errors. Steady state is defined by change in V O ₂ and V CO ₂ of <10% over a period of 5 consecutive minutes. The values for the mean REE from this steady-state period may be used as an accurate representation of the 24-hour TEE in patients with low levels of PA. In a patient who fails to achieve steady state and is metabolically unstable, prolonged testing is required (minimum of 60 minutes) and 24-hour indirect calorimetry should be considered.

Indirect calorimetry is not accurate in the setting of air leaks around the endotracheal tube, in the ventilator circuit or through a chest tube, or in patients on ECMO. A high inspired oxygen fraction ( Fi O ₂ >0.6) will also affect indirect calorimetry. Indirect calorimetry is difficult to use in babies on ECMO because a large proportion of the patient’s oxygenation and ventilation is performed through the membrane oxygenator. The use of indirect calorimetry for assessment and monitoring of nutrition intake requires attention to its limitations and expertise in the interpretation, as well as specialized equipment and personnel. Nonetheless, its application in children at high risk for underfeeding and overfeeding can be helpful.

Nonradioactive stable isotope techniques have been used to measure REE in the pediatric patient. Stable isotope technology has been available for many years and has been crucial in the study of many metabolic pathways. It was first applied for energy expenditure measurement in humans in 1982. The highly sensitive techniques of quantifying stable isotopes minimize measurement error, but the high cost of the isotopes and specialized equipment has led to limited clinical use.

In general, any increase in energy expenditure during illness or after an operation is variable, and studies suggest that the increase is far less than originally hypothesized. Newborns undergoing major surgery have only a transient 20% increase in energy expenditure that returns to baseline values within 12 hours postoperatively, provided no major complications develop. In one study, REE measurements immediately postoperatively in children with single-ventricle heart defects who underwent a Fontan procedure found a low prevalence of hypermetabolism. In another study, stable extubated neonates, 5 days after operation, were shown to have REE comparable to normal infants. Effective anesthetic and analgesic management may play a significant role in muting the stress response of the surgical neonate. A retrospective stratification of surgical infants into low- and high-stress cohorts based on the severity of underlying illness found that infants under high stress experience moderate short-term elevations in energy expenditure after operation, whereas infants under low stress do not manifest any increase in energy expenditures during the course of illness. Finally, by using stable isotopic methods, it has been found that the mean energy expenditures of critically ill neonates on ECMO are nearly identical to age- and diet-matched stable surgical neonates.

These studies suggest that critically ill neonates have only a small and usually short-term increase in energy expenditure. Although children have increased energy requirements from the increased metabolic turnover during illness, their overall caloric needs may be lower than previously thought due to possible halted or slowed growth and the use of sedation and muscle paralysis. This could result in overfeeding when energy intake is based on presumed or estimated energy expenditure with stress factors. On the other hand, unrecognized hypermetabolism in select individuals results in underfeeding with negative nutritional consequences. The variability in energy requirements may result in cumulative energy imbalances in the intensive care unit (ICU) over time.

For practical purposes, the recommended dietary caloric intake for healthy children may represent a reasonable starting point for the upper limit of caloric allotment for hospitalized children. However, as discussed earlier, energy requirement estimates in select groups of patients remain variable and possibly overestimated, mandating an accurate estimation using measured energy expenditure where available. Regular anthropometric measurements plotted on a growth chart to assess the adequacy of caloric provision will allow relatively prompt detection of underfeeding or overfeeding in most cases. However, some critically ill children may be too sick for regular weights or have changes in body water that make anthropometric measurements unreliable. The Tight Calorie Control Study (TICACOS) showed that nutritional support guided by repeated indirect calorimetry measurements in mechanically ventilated adults resulted in more frequent achievement of energy goals with higher protein delivered and a trend to lower mortality. A pediatric trial of PN titrated to measured REE in children after hematopoietic stem cell transplantation did not lead to differences in body composition. Further study into the potential benefit of nutritional delivery guided by serial measures of energy expenditure in children is warranted. V CO ₂ -based REE prediction, which relies on more widely available equipment for bedside monitoring, may make continuous metabolic assessment in mechanically ventilated patients more feasible.

Macronutrient Intake