This is the multi-professional critical care review course for pediatrics. I'll be discussing nutritional therapy in pediatric critical care setting. My name is Katri Tippo. I am division chief of pediatric critical care at University of Arizona and immediate past chair for the PICU NutriNet subgroup of Polisi. My research interests are in enteral nutrition and the gut microbiome. These slides were originally developed by Nilesh Mehta from Boston Children's Hospital and have been edited and updated by myself. I have one disclosure. I've acted as a consultant to leading biosciences for study design. The learning objectives today are to understand the metabolic stress response, understand methods for estimating energy expenditure, understand protein balance and malnutrition in the pediatric intensive care unit, and provide a pragmatic approach to bedside nutrient delivery and best practices. Let's start with the metabolic stress response. In response to systemic inflammation and critical illness, the metabolic stress response has adaptation to provide fuel to the brain in the form of lipolysis and ketone production. The negative consequence of this adaptation can be an increase in triglycerides and in the near short term, approximately within two weeks of ICU hospitalization, essential fatty acid deficiency. There is also muscle protein breakdown to create a free pool of amino acids for the liver to produce an increase in acute phase proteins. The negative consequences of this muscle protein breakdown is a decrease in nutrient transport proteins, decrease in lean muscle mass, and downstream decrease in cardiac and respiratory muscle function. Another adaptation is glycolysis and gluconeogenesis to increase glucose in the bloodstream. This results in hyperglycemia, which acts as fuel for red blood cells and the kidneys, but can have the negative consequence of impaired wound healing and immune function. Within the intestinal tract, there is decreased protein turnover, decreased glutamine utilization. There is also the dysbiosis that occurs with critical illness and loss of intestinal barrier functions. In summary, the protein catabolism after major stress and surgery is protein breakdown for redirection of amino acids for the short term adaptation to stress and an increase in acute phase proteins. But then this can have consequences when the stress response is prolonged. When we compare starvation versus metabolic stress response, look at energy expenditure in starvation, energy expenditure is decreased in the metabolic stress response. Classically, we were taught that metabolic stress increases energy expenditure, but I'll show you data later in the talk that demonstrates that this is not the case. When we evaluate hormone counter-regulatory capacity, this is preserved in starvation, but altered during the stress response with elevations in serum levels of catecholamines, cortisol, glucagon, and growth hormone. The primary fuels differ as well. In starvation, the primary fuel is fat, whereas in the stress response, the primary fuels are amino acids, glucose, and triglycerides. Ketone use is markedly increased in starvation, but only mildly so in stress response. Protein breakdown is mildly increased in the starvation, but greatly increased in the stress response. When we talk about loss of body stores of macro and micronutrients, this process is gradual in starvation, but rapid in the stress response. Gluconeogenesis and hepatic protein synthesis are mildly increased in starvation, but greatly increased in the stress response. When we talk about organ function in starvation, cardiac output, respiratory drive, GI function are all decreased. In the stress response, there is variable response, as you all know, in organ function. What happens when we provide energy and protein in starvation? There is reversal of the loss of body stores, but in the stress response, provision of energy and protein does not reverse gluconeogenesis or protein breakdown. Next, we'll talk about protein balance during the metabolic stress response. Whole body protein turnover is measuring protein synthesis, breakdown, and establishing protein balance. In health, synthesis exceeds breakdown for a positive net protein balance. In various illness states, such as measles, various infections, cancers, acute malaria, meningococcal septic shock, and thoracic surgery, while overall protein synthesis is markedly increased from healthy states, is that the breakdown is increased to an even greater degree. The net effect of this increased protein synthesis and increased protein breakdown is a negative protein balance, which results in skeletal muscle breakdown. There are special populations of critically ill children with regard to muscle catabolism after burn injury. Six, nine, and 12 months out, patients continue to have a net negative protein balance. And then in term, preterm, and older infants, they have limited stores. And so within two weeks of admission can rapidly develop cumulative protein deficits with inadequate provision of enteral nutrition. These are the SCCM and ASPA nutrition guidelines published in 2017 and recommend a minimum protein intake of 1.5 grams per kilo per day. They do not recommend RDA values, which are designed for healthy outpatient children. There is an emphasis on provision of proteins by the enteral rather than parenteral route, but it is important to note that the optimal protein dose associated with improved clinical outcomes is not known. We're going to take a trip back to undergrad or first year medical school to go over the principles underlying substrate utilization and energy production. Substrate oxidation is how energy is produced in humans. So any of the carbon-based nutrients such as carbohydrates, protein, and fat are oxidized to carbon dioxide, water, and heat. I'm sure you remember this from biochemistry. Direct calorimetry is a method used to measure this heat generation. And indirect calorimetry is a method where we measure oxygen consumption and CO2 production. And we use those values with the modified Weir equation to estimate resting energy expenditure. And respiratory quotient is the ratio of carbon dioxide production over oxygen consumption. The respiratory quotient can be used to determine combustion of different substrates based on their properties. So carbohydrates produce an RQ of 1, lipid of 0.7, protein of 0.8. And in the setting of lipogenesis due to carbohydrate overfeeding, the RQ will be greater than 1. So RQ essentially measures net substrate use. Less than 0.7 suggests the patient is underfed. And like I said, RQ greater than 1 suggests carbohydrate overfeeding. And this is important because in the setting of excess carbohydrate feeding, this can cause excess CO2 production and contribute to difficulty liberating from mechanical ventilation. In terms of energy production, carbohydrate as a substrate is 4 kilocalories per gram, whereas dextrose, commonly used in the ICU, is 3.4 kcals per gram. Fats are 9 kcals per gram, and protein is 4 kcals per gram as well. So here's our first question. Which of the following statements accurately describes nutrient metabolism? 1. Carbohydrates provide 9 kcals of energy per gram. 2. A respiratory quotient of greater than 1.2 may indicate carbohydrate overfeeding. 3. Energy burden of the metabolic stress response can be accurately estimated by equations. And 4. The metabolic stress response is characterized by increased protein breakdown and decreased protein synthesis. The correct answer is a respiratory quotient greater than 1.2 may indicate carbohydrate overfeeding. In terms of answer 1, carbohydrates actually provide 4 kcals of energy per gram. And number 3, the answer is incorrect because in fact energy burden of the metabolic stress response cannot be accurately estimated by equations. They are prone to error. And then 4. The metabolic stress response is characterized by increased protein breakdown and increased protein synthesis. We'll go next to estimating energy expenditure and the role of indirect calorimetry. So how to estimate resting energy expenditure? In this prospective observational study in critically ill children, REE was measured by indirect calorimetry. Somewhat surprisingly, at least at the time when this study was published, REE was not influenced by severity of illness as measured by PRISM-PIM2 or PILADS score, diagnostic category, baseline nutritional status, or biochemical status. And when we compare accuracy of equation-estimated resting energy expenditure to indirect calorimetry in a study with 52 critically ill children, none of the existing equations, and they compared Harris-Benedict, WHO, Mayo, Talbot, Schofield, and RDA, perform well to predict within 10% of measured resting energy expenditure. The best performs only well about 40% of the time. And so none of the existing equations accurately assess measured resting energy expenditure. And the inaccuracy of equation-estimated resting energy expenditure is repeatedly demonstrated in the literature and is made worse, in fact, with the addition of stress factors added to the equation to estimate resting energy expenditure. Indirect calorimetry is the preferred method to estimate resting energy expenditure in critically ill patients. It actually measures the respiratory exchange ratio, which is an estimate of RQ, which, as a reminder, is carbon dioxide production over oxygen consumption. And in ideal testing conditions at steady state, RER approximates respiratory quotient. So how do we do this? You can, in the diagram here on the left, it's an intubated patient as an example, and there's a sensor at the endotracheal tube and there's another sensor at the expiration gas valve on the ventilator. And then you look at your measuring CO2 production and oxygen consumption in a non-intubated patient, as you see in the picture, that's me and our PICU. You can put a bubble over the patient with bias flow, and then similarly it measures CO2 production and oxygen consumption. So the WEIR equation, we take measured values of carbon dioxide at the endotracheal tube and at the ventilator, expiratory port, and oxygen levels, and then we calculate respiratory quotient and resting energy expenditure using this WEIR equation. And indirect calorimetry measures resting energy expenditure, not total daily energy expenditure, but as you can imagine in a mechanically ventilated sedated patient, for example, the two total energy expenditure and resting are fairly good approximations of each other. In an outpatient, resting energy expenditure is about 60 to 80 percent of total energy expenditure. 10 percent is SDA or diet-induced thermogenesis. 10 to 30 percent, depending on activity level, is physical activity energy expenditure. And in children, there is the additional energy expenditure used for growth. When we look at resting energy expenditure in critically ill children, it is actually more often hypometabolism as opposed to what had previously been hypothesized, that most critically ill children and adults were hypermetabolic. This turns out not to be the case when we actually measure resting energy expenditure. In the graph to the left, this is resting energy expenditure after hematopoietic stem cell transplantation and time from admission in weeks along the x-axis. And along the y-axis is percent of predicted basal metabolic rate. So this is basal metabolic rate. So patients are below basal metabolic rate post stem cell transplant for several weeks. And then the second graph on the right is resting energy expenditure measured eight hours after Fontan palliation with cardiopulmonary bypass. And in this graph, you can see that between patients, there is variability. And in an unpredictable pattern, you can't identify which patients, based on biochemical parameters, are going to be hyponormal or hypermetabolic. And so in this graph, in this table, there are nine patients who are hypometabolic, 10 who are normometabolic, and 7 who are hypermetabolic. But the predominance is normal or hypometabolic. So this concept that critically ill children with systemic inflammation are hypermetabolic, it does not bear out when patients' metabolism is measured. Not only is there between patient variability and resting energy expenditure, there is patient variability over the course of critical illness. This is a single patient with multiple metabolic CART tests done over the course of their prolonged hospitalization after liver transplant. And if you look at the y-axis, it's kcals per kilo per day. And if you look at the totals, they are very modest. So the maximum kcals per kilo per day measured was 45 kcals per kilo per day. And in this one spot here where the patient became septic, actually only needing 20 kcals per kilo per day. So quite hypometabolic. So when we evaluate overall energy intake in the PICU or energy prescription at goal delivery, patients who are in the PICU, there's both interpatient variability and within-patient variability. So at any given time, a patient may be at risk of underfeeding if they happen to be transiently hypermetabolic, or at other times at risk of overfeeding, actually the majority of the time. And this energy imbalance may have unintended consequences. So this is the classic Goldilocks dilemma, right? Energy delivery needs to be just right. And actually, energy overfeeding is quite common. So if you look at this graph of resting metabolism across days for major burn, major trauma, minor trauma patients, you can imagine that if the calorie goal is set early on and never changed, even if it's set to a measured energy expenditure, that later in the hospital stay, you are accomplishing unintended overfeeding, which has negative consequences. In this study from J. Penn in 2015, they measured indirect calorimetry repeatedly in patients in the pediatric intensive care unit. And what they discovered was the percent of days overfed, underfed, or adequate feeding. And so 60% of days, patients were overfed, 20% underfed, and 20% were adequately fed, meaning that energy delivery met measured energy expenditure. And so this speaks to the concept that the energy prescription should change over time in response to measured needs. In burn patients, this is a special class of patients where they are predictably and reliably hyper-metabolic post-burn. The resting energy expenditure is elevated for months after the injury. And this is the patient population where the old hypothesis that critical illness induces hyper-metabolism is supported. So what do the guidelines say? These are the SCCM and ASPEN nutrition guidelines published in 2017. Their recommendation is that measured energy expenditure by indirect calorimetry is the preferred method to determine energy requirements. If IC measurement is not possible, then use of Schofield or WHO estimating equations without the addition of any stress factors to estimate energy expenditures. And then when they determine the target energy goal, the recommendation is to achieve delivery of at least two-thirds of the prescribed daily energy requirement by the end of the first week in the PICU. The rationale for that recommendation is multiple observational studies demonstrating that that's sort of the cutoff point for improved outcomes. So when we look at energy balance in a critically ill patient, there are elements of support provided in the ICU that decrease demands in terms of energy requirements, such as mechanical ventilation, muscle relaxants, sedatives, and passive feeding. And then there are some items that may temporarily stress or increase the energy demands for the patient, such as surgery, illness procedures. And this is not uniform because things like an intercurrent sepsis that's hospital-acquired may actually reduce energy requirements as opposed to increasing them. But a patient with a burn injury, for example, will have elevated and persistent injury. And if a patient has surgery and requires healing, they may be temporarily hyper-metabolic. So energy balance becomes important over time and because it relates to patient outcomes. So overfeeding is associated with a net lipogenesis, hepatic steatosis, increased carbon dioxide production, which may impact liberation from mechanical ventilation, that provides an increased ventilatory burden, and may be associated with hyperglycemia. And as a consequence, this may cause delayed weaning from mechanical ventilation. In turn, underfeeding is associated with malnutrition, failure to meet protein and calorie requirements, and loss of lean body mass, which then can also cause a delay in weaning from mechanical ventilation. So our next question, which statements accurately describe energy expenditure during critical illness? So one is resting energy expenditure can be accurately estimated using the Schofield equation. Two, the response to starvation is characterized by elevated energy expenditure. Three, overfeeding imposes an increased carbon dioxide burden. Four, energy expenditure may be increased for weeks after severe burn injury. The correct answer is overfeeding imposes an increased carbon dioxide burden. As we learned in the prior slides, resting energy expenditure cannot be accurately estimated using the Schofield equation, even though that is the equation that's currently recommended in the guidelines. The response to starvation, answer two, is characterized by decreased energy expenditure. And then four, energy expenditure may be increased, it's actually for months after severe burn injury. Next, we'll look at protein balance and preservation of lean body mass. This is the first of a couple of studies that I'll show you regarding protein balance in the PICU. In this systematic review, they identified that positive protein balance required provision of protein intake of more than 1.5 grams per kilo per day. You see this in the leftward graph where balance is in the y-axis, so a positive number is on top. And then this cut point with the blue hash line, really where you see the first patients with positive protein balance are greater than 1.5 grams per kilo per day. But you also need adequate energy intake in order to achieve positive protein balance because you need appropriate fuel sources. So in this study here, again, on the right graph, there's kcals per kilo per day on the x-axis and protein balance on the y. So the top half of the graph are patients where they had positive protein balance, and the cutoff point in terms of calories was 57 kcals per kilo per day. This is a study from Michael August and from 2006. Their question really was to address protein balance in response to insulin. But what they found is relevant to our conversation because in the presence of inadequate nutrition, not only was there no response to insulin in terms of improving protein balance, but there was overall low protein synthesis. But in the setting of adequate nutrition, there was improvement in overall protein synthesis and protein balance, but then patients actually had a response to insulin in the form of increased protein synthesis. And similar to other studies, they identified that improvement in protein balance was due to increased synthesis rather than reduced breakdown. And then we don't do a very good job delivering adequate protein intake in pediatric ICU patients. In this graph, it's days in PICU on the x-axis and enteral protein adequacy on the y-axis. And what you see is that on day seven, the median and mean intake are 40% and 50% of goal. And so patients are developing protein deficits over the first week of ICU stay. So does it matter, right? It does. So when you look at protein intake and outcomes, survivors are more likely to achieve improved protein adequacy as compared to non-survivors. And then looking at an incremental effect of protein adequacy on the probability of mortality on the graph on the right, you can see that with improving protein adequacy, at every 5%, you see an improved, decreased probability of mortality in these patients. Factors that predict optimal protein delivery in the PICU include time to initiating enteral nutrition after admission in days, the route of enteral nutrition delivery, total duration of enteral nutrition interruption in hours, the presence of a dedicated PICU dietician, and the number of PICU beds in your unit, presumably a surrogate for volume. And when we think about postoperative patients, how could we achieve postoperative anabolism? So we can do that by providing optimal protein and energy to patients. This is a study of pediatric cardiac surgery patients on post-operative days three to 10. And similar to the prior systematic review where they identified cut points for a shift to positive protein balance or anabolism. In this study, they identified a minimum intake of 55 kcals per kilo per day and one gram protein per kilo per day. And that that was associated with anabolism or a positive protein balance. Now I'll shift to discussion of malnutrition specifically focused on hospital acquired macronutrient malnutrition and micronutrient deficiencies. This is the UNICEF global malnutrition map with the key on the left side where countries in the pink color have a greater than 40% or very high baseline malnutrition risk. And then from the turquoise to aqua to different greens, you get to the countries with the lowest prevalence of baseline malnutrition. And what I've done is superimpose these blue circles describing the reported prevalence of malnutrition in the Pediatric Intensive Care Unit. So even in countries with baseline low malnutrition, our patient population has a significant burden of malnutrition and worldwide, the malnutrition in the Pediatric Intensive Care Unit is about 30%. So why do patients in the PICU acquire malnutrition? So as we've just described previously, they have minimal substrate reserves and altered metabolism. And then while they're in the PICU, we don't deliver adequate nutrition. And then there's an interplay with potential baseline chronic illnesses for patients in the PICU. This all contributes to the data of 30% of patients having preexisting malnutrition and then 66% of patients being malnourished at discharge. So they've acquired malnutrition while in the PICU. There are many micronutrient deficiencies that have been described in pediatric critical illness such as selenium and vitamin C, which in some literature is associated with impaired wound healing and immune dysfunction, vitamin D deficiency with an association with need for increase in catecholamine infusions, additional fluid resuscitation, mechanical ventilation, and PICU length of stay in isolated studies of vitamin D deficiency. Zinc, copper and iron are associated with multi-organ failure, increased severity of illness and impaired wound healing. And thiamine is associated with neurological and cardiovascular dysfunction. The issue is really that these low micronutrient concentrations do not necessarily reflect deficiency states. They may reflect redistribution or an adaptive response to critical illness. And then dietary reference intakes are estimated from healthy children and so likely don't bear relevance to the critically ill patient population. So provision of micronutrients at pharmacologic doses has been associated with no effect or harm in all of the studies where they've been used. So the current recommendation is to provide micronutrients at reference intake requirements for age. And clearly more studies are needed to understand micronutrient needs during pediatric critical illness. When there have been studies to look at pharmaconutrition, specifically looking at immune modulating nutrients, in adult ICU, the reducing deaths due to oxidative stress or the redox study, used glutamine and or antioxidants, such as the selenium, zinc, beta carotene, vitamin E and C, and they looked at a primary outcome of 28 day mortality and did not identify a benefit and there was some signal for harm. And then in the PICU, the randomized comparative pediatric critical illness stress-induced immune suppression or crisis prevention trial, studied enteral zinc, selenium, glutamine and IV metoclopramide with a primary outcome of time-to-nose colonial sepsis or infection, and it was stopped early due to futility as there was no benefit. So in the current surviving sepsis pediatric guidelines for 2020, with regard to micronutrient, vitamin supplementation and sepsis, these are the recommendations. So they suggest against the use of glutamine supplementation in children with septic shock or organ dysfunction. They also suggest against the use of arginine, against the use of zinc supplementation, against the use of ascorbic acid, against the use of thiamine to treat children with sepsis associated organ dysfunction, and they suggest against the acute repletion of vitamin D deficiency for treatment of septic shock or organ dysfunction. So I hope kind of something that you've gleaned already is that nutritional assessment in the ICU has many technical challenges. Not only is it difficult to correctly identify energy and protein requirements, but looking at nutritional status at time of admission and throughout the stay is very complex. There's a combination of anthropometrics that people have used such as weight, skinfold thickness and biochemical data. People have used dry weight, weight percentiles or standardized weights with C-scores, various growth charts or water low criteria for acute or chronic malnutrition. The issue in the ICU, as you all know, is that these anthropometric measurements are often altered secondary to capillary leak syndrome or edema and weight gain. When we assess metrics typically used in the outpatient setting to assess nutritional status, they often don't function well in the pediatric ICU. So weight versus standardized curves is confounded by the presence of edema in the pediatric intensive care unit. Height would be wonderful for longer stay patients if it were reliably collected, but so often it is either not collected or it is inaccurately assessed. Lymphocyte count is often used to assess immune function, but in the ICU, as you know, it's falsely low or high. Similarly, hemoglobin as an assessment of iron status is often falsely low in the ICU. Energy panels to assess immune function may be confounded by medications provided in the ICU. Albumin is a visceral protein marker of nutritional status, but in the acute critical illness setting is affected by capillary leak and systemic inflammation driving immune transport. And then albumin and retinol binding protein are also visceral proteins used to assess nutritional status but are confounded by hepatic function. So assessments of the causes and consequences of pediatric nutrition is quite complex. We have anthropometry parameters such as weight height or length, skin folds, mid-upper arm circumference, which incidentally mid-upper arm circumference is actually the most resistant measure to fluid shifts in edema. And then we have Z-scores to normalize these values and reference charts provided by WHO or CDC. Then when looking at the etiology and chronicity of malnutrition, malnutrition can be non-illness related or it can be illness related and then acute or chronic. And then it can be further modified by the presence of systemic inflammation. And then when we look at mechanisms for malnutrition, there are starvation with anorexia or due to socioeconomic factors, iatrogenic feeding interruptions or intolerance to feeds. And then in terms of illness related mechanisms, there can be malabsorption, nutrient losses, hypermetabolism due to increased energy expenditures, for example, in cystic fibrosis patients or patients with chronic lung disease. And then during inflammation, you can further have altered utilization of nutrients. And so then you have this imbalance of nutrients with potentially decreased intake or increased nutrient requirements leading to malnutrition. And for both macronutrients, so protein and calories, but also potentially micronutrient deficiencies. And so the outcomes of interest are really loss of lean body mass, muscle weakness, in the long-term developmental or intellectual delays, risk for infection or immune dysfunction, delayed wound healing in postoperative patients, and ultimately prolonged hospital length of stay. So one of the issues with serum proteins as biomarkers of nutritional status is that in acute illness, they are often altered. So we have positive acute phase proteins and negative acute phase proteins. So the positive acute phase proteins you're all familiar with, CRP, fibrinogen, ferritin, seroplasmin, alpha-1-antripsin, and alpha-1-glycoprotein. But negative acute phase proteins, you notice that these are all on the list for assessment of nutritional status. So albumin, transferrin, prealbumin, and retinol-binding protein, plasma concentrations are all decreased as a result of acute inflammation and acute critical illness, so may not reflect nutritional status because they've shifted into another compartment. So we can look at the half-lives of visceral protein markers of nutritional status, which once systemic inflammation is resolving and the patient's status is improving, the plasma concentrations may normalize and begin to reflect nutritional status. And so albumin has a half-life of 20 days, transferrin, eight days, prealbumin, two days, and retinol-binding protein, 12 hours. Next, we'll look at pragmatic approaches to bedside nutrient delivery and assess best practices and impact on clinical outcomes. First of all, touch on specific guidelines for preoperative fasting in children, looking at various societies that have released guidelines. The American Academy of Pediatrics Committee on Drugs recommends clear fluids up to two hours prior to anesthesia and then milk or solids up to four hours for neonates, six hours for infants, and eight hours for children prior to surgery. And the American Society of Anesthesiologists recommends a six-hour fast from non-human milk or infant formula for neonates and infants and four-hour fast from breast milk preoperatively. The Royal College of Nursing from 2005 essentially follow the ASA recommendations of the 2-4-6 rule, so two hours for clear fluids, four hours for breast milk, and six hours for solids, including formula or cow's milk as counted as a solid. And the Canadian Anesthesiologist Society from 2008, they also follow the 2-4-6 rule, but state eight hours after a meal that includes meat, fried, or fatty foods. The Scandinavian guidelines from 2005, they differ only in their inclusion of formula milk in the four-hour rule along with breast milk. Based on these recommendations, really a prudent preoperative nutrition strategy would be to avoid prolonged fasting and to follow ASA guidelines for healthy, not at-risk patients using the 2-4-6 rule. So fluids at two to three hours prior to surgery in prior studies do not influence gastric volume, pH, emptying, or risk of aspiration. In high-risk children, which means patients who have obesity or dysmotility or ileus or who have been on opioid infusions, you have to exercise caution because there really is no evidence to guide the strategy for these patients. Interestingly, there's data to suggest that carbohydrate loading can shift a patient's metabolic status from the fasted to fed state. So providing 50 grams of carbohydrate entrally two hours prior to surgery has been shown to decrease insulin resistance, modulate post-operative stress response favorably, decrease protein catabolism, muscle loss, and preserve muscle function post-op and may reduce hospital post-operative stay. Now we'll turn to the route of enteral feeds comparing gastric versus post-pyloric or small bowel feeding. This is an RCT of 74 ICU patients by Kathy Meert. They compared continuous feeds provided by the gastric or small bowel route. Of note, daily calorie goal achieved was significantly lower in the gastric group versus the small bowel group, but there were no differences identified in microaspiration, tube displacement, and intolerance between the groups. Now of note, this was not powered to detect mortality and it was an inadequate tool to detect aspiration. Notably, feeding tube placement was not achieved in 12 patients in the small bowel group. So the group analyzed was 12 patients smaller in the small bowel group and then feeds were advanced at the discretion of nursing staff. When we look at continuous versus bolus feeds, there's not a whole lot of literature on this topic, but just recently, Emory Brown published in J-PEN in 2021 comparing patients with bolus versus continuous feeds. They found that patients in the bolus fed group had a greater percentage of goal calories and protein delivered during the study time period. There were fewer interruptions to feeds in the bolus fed group. They included a randomization strategy to look at intolerance metrics. So they had a group with MS as an only indicator of intolerance versus allowing gastric residual volume measurement. The majority of interruptions throughout the study were due to gastric residual volume being elevated. And it is really interesting that in fact, no MS's event was preceded by an elevated GRV and that's supported, well supported by the adult literature where they're turning away from gastric residual volume assessments as a way to assess intolerance. So how do we improve macronutrient intake in pediatric critical illness? The answer is fairly straightforward actually is to implement, the best way is really to implement a stepwise nutrition delivery algorithm in your PICU. This study from Hamilton in 2014 demonstrated comparing a pre-intervention and post-intervention cohort identified implementation of a nutrition guideline improving the proportion of patients receiving their energy goal in the days after PICU admission. This is sort of the culmination of an assessment of patient requirements, making a recommendation, providing prescription and providing appropriate delivery of nutrition. Now we'll talk about parenteral nutrition. So the PAPANIC trial was an early versus late trial of parental nutrition in critically ill children. Parental nutrition was started within 24 hours if enteral nutrition did not reach 80% of goal calories. And they compared that to parenteral nutrition started on day eight. And these are the calorie graphs for patients enrolled in the study. The top set of graphs are for zero to 10 kilogram patients and the Y-axis is K-cals per kilo per day. One thing to note is that, so looking at the leftmost upper graph, you see enteral nutrition in both randomized groups by day four was about 40 K-cals per kilo per day. Then parenteral nutrition in addition to that was also by day four at about 50 K-cals per kilo per day. And so the total calories provided by additional parental nutrition, the total combined was at about by day four, 80 to 90 K-cals per kilo per day, which you remember prior calorie targets in the description of hypometabolism in critically ill children, this may represent overfeeding. And the similar pattern was seen in patients 10 to 20 kilos in size, which is the lower series of graphs. So this excellent study of early versus late PN in critically ill children, really is comparing very early versus late or no parenteral nutrition. And the question is really, does this represent overfeeding as a comparison versus appropriate feeding? And what did they find? So in the late parenteral nutrition group, they identified 7.8% lower rate of new infection defined as an airway or bloodstream infection. They had patients had a shorter PICU length of stay by 2.7 days and a higher likelihood of discharge alive. The message here is really to avoid unnecessary parenteral nutrition administration in the early days of PICU hospitalization to avoid overfeeding. There remain many questions regarding provision of IV micronutrients, but ultimately the goal remains to advance enteral nutrition as tolerated using a stepwise algorithm. I think most of you are familiar with parenteral nutrition complications. So first and foremost is infection or CLABSI and potential strategies to mitigate this are precautions while placing central line and subsequent access of the line. And then in parenteral nutrition associated labor disease in chronically PN dependent patients, there are multiple strategies to try to mitigate this problem. So ethanol locks, cycling PN, soy oil emulsions restricting to one gram per kilo per day or fish oil emulsions, which are currently only available as compassionate use in the United States. And then multidisciplinary intestinal rehabilitation programs. And then in terms of avoiding phlebitis is restricting osmolarity to 900 millisieverts per liter. An overall nutritional strategy should really be to use the gut when it works, early EN initiation and protocolized advancement and PN only when EN is not tolerated or not able to meet nutrient delivery goals either due to chronic intolerance or short bowel. And then the timing to switch to parenteral nutrition is really can be made based on the patient's baseline nutritional status and their overall nutritional risk. If you do need to use PN, planning a PN prescription involves determining energy requirements measured versus estimated and then planning a nutrient composition. So generally lipids will be 25 to 40% of calories protein, 10 to 20% carbohydrates as 45 to 60%, typically starting with about 10% dextrose and advancing to goal gradually and then using an electrolytes trace elements and a multivitamin mixture. And then how to calculate the energy content of PN is lipids are calculated as 10 kcals per gram, protein is four and dextrose is 3.4. This is a topic that I have a lot of interest in enteral nutrition and cardiovascular meds in the PICU. So this is a study from Dr. King, published in J-PEN in 2004, looking at E intolerance for patients on vasoactive meds, ages one month to 20 years of age. In 29% of patients, enteral nutrition was interrupted for intolerance. And then the important sort of outcome was really GI bleeding, four out of 52 patients experienced this, but it was deemed not to be related to feeds and not clinically significant. And their conclusion was really that pediatric patients receiving cardiovascular medications tolerate enteral nutrition without adverse events. This is a retrospective study from Penchell in 2016, using NACRI focus group data, comparing patients who were on, by the vasoactive inotrope score, by whether they were fed versus not fed. Their findings were that only on day one was there a difference in vasoactive inotrope score between the fed and not fed groups of patients. And notably the median VIS score for patients who were fed was 10, which if you recall, represents a dopamine or dibutamine dose of 10 mics per kilo per minute, or an epi and or epi of 0.1. Importantly, they did not identify any adverse GI outcomes between the fed and not fed groups. And they assessed vomiting, diarrhea, GI bleeding, abdominal distension. The only patient to be diagnosed with necrotizing enterocolitis was in the non-fed group. And obviously only patients who were being fed had metrics of feeding intolerance. What does the surviving sepsis guideline from 2020 say about enteral nutrition? The recommendation is to commence early enteral nutrition within 48 hours of admission in children with septic shock who have no contraindications to enteral nutrition, and to increase enteral nutrition in a stepwise fashion until nutrient goals are met. So to review further their guidelines from 2020, so they suggest enteral nutrition as the preferred method of feeding, and that parental nutrition be withheld for the first seven days of PICU admission. This is based on the PAPANIC data. They suggest against supplementation with specialized lipid emulsions, and against the routine measurement of gastric residual volume. They also suggest administering enteral feeds through a gastric tube rather than post-pyloric feeding, and that the routine use of prokaryotic feed and that the routine use of prokinetic agents for the treatment of feeding intolerance is not recommended. So another question, the main advantages of enteral over parenteral nutrition include one, preservation of gastrointestinal function and mucosal integrity, two, lower cost, three, decreased likelihood of acquired infections, and the answer is all of the above. A pragmatic approach to natural nutrition is really to identify the vulnerable at-risk patients, identify patients with poor baseline nutritional status, provide an individualized energy and protein prescription to initiate early enteral nutrition within 24 to 48 hours after admission, after hemodynamic stability is achieved, and then cautious protocol and guidelines for advancing enteral nutrition and monitoring for intolerance. Really the use of gastric residual volumes remains a question as there haven't been a large number of studies in the PICU, but its utilization is certainly being called into question and then to stop if there are concerns regarding intestinal compromise and restart the feeds when that concern is abated. Nutrition should really be included as part of discussions on daily rounds with identified nutrition goals, and there's no role for early parenteral nutrition within the first few days. There is no magic fits all therapy for a heterogeneous population, and then we need to follow up the patients to document meaningful outcomes. So looking back at our learning objectives, so we reviewed metabolic stress response and learned that the protein catabolism energy burden cannot be easily estimated. We've looked at role of indirect calorimetry to target at high-risk patients, looked at protein balance and the importance of preserving lean body mass, looked at malnutrition and hospital-acquired malnutrition, and looked at best practices in terms of barriers, underfeeding and overfeeding risk, mortality and morbidity, and then using an algorithmic approach, having nutrition support team and dedicated dietician. In summary, children have unique metabolic responses to trauma, surgery and burns. Energy expenditure is not increased as was previously postulated. Protein catabolism is of paramount importance. Wound healing, lean mass and function loss and outcomes need to be assessed. Protein intake must be optimized, but does need further investigation. Internal nutrition is challenging, but it is feasible in most patients. And the role of universal early and aggressive pre-nutrition is certainly questioned at this point. Non-nutritional strategies to prevent muscle wasting are emerging and that early mobilization, physical therapy and pharmacotherapy may be able to assist in preserving lean muscle mass. I would like to thank you for your time and have included my email address for any questions.

nutritional therapy

pediatric critical care

metabolic stress response

lipolysis

ketone production

muscle protein breakdown

glycolysis

triglycerides