Journal of Parenteral and Enteral            Nutrition                   A.S.P.E.N. Clini...
Journal of Parenteral and                                                                                                 ...
A.S.P.E.N. Clinical Guidelines / Mehta et al   261an ongoing basis, reviewing about 20% of the chapters                   ...
262   Journal of Parenteral and Enteral Nutrition / Vol. 33, No. 3, May/June 2009                                         ...
A.S.P.E.N. Clinical Guidelines / Mehta et al   263proportional to the severity of injury. Many hospitals are              ...
264   Journal of Parenteral and Enteral Nutrition / Vol. 33, No. 3, May/June 2009                                         ...
A.S.P.E.N. Clinical Guidelines / Mehta et al   265                                                    Table 4             ...
266   Journal of Parenteral and Enteral Nutrition / Vol. 33, No. 3, May/June 2009                                         ...
A.S.P.E.N. Clinical Guidelines / Mehta et al   267underfeeding in selected subjects, and the cost-benefit                 ...
268   Journal of Parenteral and Enteral Nutrition / Vol. 33, No. 3, May/June 2009Glycogen stores are limited and quickly d...
A.S.P.E.N. Clinical Guidelines / Mehta et al   269children receiving early (< 24 hours after PICU admission)              ...
270   Journal of Parenteral and Enteral Nutrition / Vol. 33, No. 3, May/June 2009                                         ...
A.S.P.E.N. Clinical Guidelines / Mehta et al   271                                                                 Table 5...
Aspen children
Aspen children
Aspen children
Aspen children
Aspen children
Upcoming SlideShare
Loading in...5

Aspen children


Published on

Published in: Health & Medicine, Business
1 Like
  • Be the first to comment

No Downloads
Total Views
On Slideshare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

Aspen children

  1. 1. Journal of Parenteral and Enteral Nutrition A.S.P.E.N. Clinical Guidelines: Nutrition Support of the Critically Ill Child Nilesh M. Mehta, Charlene Compher and A.S.P.E.N. Board of Directors JPEN J Parenter Enteral Nutr 2009 33: 260 DOI: 10.1177/0148607109333114 The online version of this article can be found at: Published by: On behalf of: The American Society for Parenteral & Enteral Nutrition Additional services and information for Journal of Parenteral and Enteral Nutrition can be found at: Email Alerts: Subscriptions: Reprints: Permissions: >> Version of Record - Apr 27, 2009 What is This? Downloaded from at International Islamic University Malaysia on February 15, 2012
  2. 2. Journal of Parenteral and Enteral Nutrition Volume 33 Number 3A.S.P.E.N. Clinical Guidelines: Nutrition May/June 2009 260-276 © 2009 American Society forSupport of the Critically Ill Child Parenteral and Enteral Nutrition 10.1177/0148607109333114 hosted at http://online.sagepub.comNilesh M. Mehta, MD, DCH1; Charlene Compher, PhD,RD, CNSD2; and A.S.P.E.N. Board of Directors Background careful selection of the appropriate mode of feeding and monitoring the success of the feeding strategy. The use ofThe prevalence of malnutrition among critically ill patients, specific nutrients, which possess a drug-like effect on theespecially those with a protracted clinical course, has immune or inflammatory state during critical illness, con-remained largely unchanged over the last 2 decades.1,2 tinues to be an exciting area of investigation. The lack ofThe profound and stereotypic metabolic response to criti- systematic research and clinical trials on various aspectscal illness and failure to provide optimal nutrition support of nutrition support in the PICU is striking and makes ittherapy during the intensive care unit (ICU) stay are the challenging to compile evidence based practice guide-principal factors contributing to malnutrition in this lines. There is an urgent need to conduct well-designed,cohort. The metabolic response to stress, injury, surgery, multicenter trials in this area of clinical practice. Theor inflammation cannot be accurately predicted and the extrapolation of data from adult critical care literature ismetabolic alterations may change during the course of not desirable and many of the interventions proposed inillness. Although nutrition support therapy cannot reverse adults will have to undergo systematic examination andor prevent this response, failure to provide optimal nutri- careful study in critically ill children prior to their appli-ents during this stage will result in exaggeration of cation in this population.existing nutrient deficiencies and in malnutrition, which In the following sections, we will discuss some of themay affect clinical outcomes. Both underfeeding and key aspects of nutrition support therapy in the PICU;overfeeding are prevalent in the pediatric intensive care examine the literature and provide best practice guidelinesunit (PICU) and may result in large energy imbalances.3 based on evidence from PICU patients, where available.Malnutrition in hospitalized children is associated with While some PICU popu­ations include neonates, lincreased physiological instability and increased resource A.S.P.E.N. Clinical Guidelines for neonates will beutilization, with the potential to influence outcome from published as a separate series.critical illness.4,5 The goal of nutrition support therapiesin this setting is to augment the short-term benefits of thepediatric stress response while minimizing the long-term Methodologyharmful consequences. Accurate assessment of energyrequirements and provision of optimal nutrition support The American Society for Parenteral and Enteral Nutri­therapy through the appropriate route is an important tion (A.S.P.E.N.) is an organization comprised of health-goal of pediatric critical care. Ultimately, an individua­ care professionals representing the disciplines of medi-lized determination of nutrient requirements must be cine, nursing, pharmacy, dietetics, and nutrition science.made to provide appropriate amounts of both macro- and The mission of A.S.P.E.N. is to improve patient care bymicronutrients for each patient at various times during advancing the science and practice of nutrition supportthe illness course. The delivery of these nutrients requires therapy. A.S.P.E.N. vigorously works to support quality patient care, education, and research in the fields of nutrition and metabolic support in all healthcare settings. These clinical guidelines were developed under the guid-From 1Critical Care Medicine, Dept. of Anesthesia, Children’sHospital, Boston, and 2University of Pennsylvania School of ance of the A.S.P.E.N. Board of Directors. Promotion ofNursing, Philadelphia. safe and effective patient care by nutrition support prac- titioners is a critical role of the A.S.P.E.N. organization.Address correspondence to: Charlene W. Compher, PhD, RD,FADA, LDN, CNSD, University of Pennsylvania School of Nursing, The A.S.P.E.N. Board of Directors has been pub­ishinglClaire M. Fagin Hall, 418 Curie Boulevard, Philadelphia, PA clinical guidelines since 1986.6-8 Starting in 2007,19104-4217; e-mail A.S.P.E.N. has been revising these clinical guidelines on260 Downloaded from at International Islamic University Malaysia on February 15, 2012
  3. 3. A.S.P.E.N. Clinical Guidelines / Mehta et al   261an ongoing basis, reviewing about 20% of the chapters Table 1each year in order to keep them as current as possible. Grading of Guidelines and Levels of Evidence These clinical guidelines were created in accordance Grading of Guidelineswith Institute of Medicine recommendations as “syste­matically developed statements to assist practitioner and A Supported by at least two level I investigationspatient decisions about appropriate health care for specific B Supported by one level I investigationclinical circumstances.”9 These clinical guidelines are for C Supported by level II investigationsuse by healthcare professionals who provide nutrition D Supported by level III investigationssupport services and offer clinical advice for managing E Supported by level IV or V evidenceadult and pediatric (including adolescent) patients in Levels of Evidenceinpatient and outpatient (ambulatory, home, and specia­ I Large randomized trials with clear-cut results; low risk oflized care) settings. The utility of the clinical guidelines is    false-positive (alpha) and/or false-negative (beta) errorattested to by the frequent citation of this document in II Small, randomized trials with uncertain results;peer-reviewed publications, and their frequent use by    moderate-to-high risk of false-positive (alpha) and/orA.S.P.E.N. members and other healthcare professionals    false-negative (beta) errorin clinical practice, academia, research, and industry. III Nonrandomized cohort with contemporaneous controlsThey guide professional clinical activities, they are help­ IV Nonrandomized cohort with historical controlsful as educational tools, and they influence institutional V Case series, uncontrolled studies, and expert opinionpractices and resource allocation.10 Reproduced from Dellinger RP, Carlet JM, Masur H. Introduction. These clinical guidelines are formatted to promote Crit Care Med. 2004;32(11)(suppl):S446 with permission of thethe ability of the end user of the document to understand publisher. Copyright 2004 Society of Critical Care Medicine.the strength of the literature used to grade each recom­mendation. Each guideline recommendation is presented Systematic reviews are a specialized type of literatureas a clinically applicable definitive statement of care and review that analyzes the results of several RCTs, and mayshould help the reader make the best patient care decision. receive a grade level of I or II, depending on the overallThe best available literature was obtained and carefully quality of the reports. Meta-analyses can be used toreviewed. Chapter author(s) completed a thorough combine the results of studies to further clarify the overallliterature review using Medline, the Cochrane Central outcome of these studies but will not be considered in theRegistry of Controlled Trials, the Cochrane Database of grading of the guideline. A level of III is given to cohortSystematic Reviews and other appropriate reference studies with contemporaneous controls, while cohortsources. These results of the literature search and review studies with historic controls will receive a level of IV.formed the basis of an evidence-based approach to the Case series, uncontrolled studies, and articles based onclinical guidelines. Chapter editors work with the authors expert opinion alone will receive a level of ensure compliance with the author’s directives regardingcontent and format. The initial draft is then reviewedinternally to ensure consistency with the other A.S.P.E.N. Practice Guidelines and RationalesGuidelines and Standards and externally reviewed (byexperts in the field within our organization and/or out­ ide s Table 2 provides the entire set of guidelines recom-of our organization) for appropriateness of content. Then mendations for nutrition support in the critically ill child.the final draft is reviewed and approved by the A.S.P.E.N.Board of Directors. The system used to categorize the level of evidence 1. Nutrition Assessmentfor each study or article used in the rationale of the guide­ 1A) Children admitted with critical illnessesline statement and to grade the guideline recommendation should undergo nutrition screening to identifyis outlined in Table 1.11 those with existing malnutrition or those who The grade of a guideline is based on the levels of are nutritionally at-risk. Grade Devidence of the studies used to support the guideline. 1B) Formal nutrition assessment with the devel-A randomized controlled trial (RCT), especially one that opment of a nutrition care plan should beis double blind in design, is considered to be the stron­ est g required, especially in those children with pre-level of evidence to support decisions regarding a therapeutic morbid malnutrition. Grade Eintervention in clinical medicine.12 A level of I, thehighest level, will be given to large RCTs where results Rationaleare clear and the risk of alpha and beta error is low (well-powered). A level of II will be given to RCTs that include The prevalence of malnutrition in hospitalized childrena relatively low number of patients or are at moderate- has remained unchanged over several years and has impli-to-high risk for alpha and beta error (under-powered). cations on hospital length of stay (LOS), illness course Downloaded from at International Islamic University Malaysia on February 15, 2012
  4. 4. 262   Journal of Parenteral and Enteral Nutrition / Vol. 33, No. 3, May/June 2009 Table 2 Nutrition Support Guideline Recommendations in the Critically Ill Child# Guideline Recommendations Grade1 1A) Children admitted with critical illnesses should undergo nutrition screening to identify those with existing D malnutrition and those who are nutritionally-at-risk. 1B) A formal nutrition assessment with the development of a nutrition care plan should be required, especially in E those children with premorbid malnutrition.2 2A) Energy expenditure should be assessed throughout the course of illness to determine the energy needs of D critically ill children. Estimates of energy expenditure using available standard equations are often unreliable. 2B) In a subgroup of patients with suspected metabolic alterations or malnutrition, accurate measurement of E energy expenditure using indirect calorimetry (IC) is desirable. If IC is not feasible or available, initial energy provision may be based on published formulas or nomograms. Attention to imbalance between energy intake and expenditure will help to prevent overfeeding and underfeeding in this population.3 There are insufficient data to make evidence-based recommendations for macronutrient intake in critically ill E children. After determination of energy needs for the critically ill child, the rational partitioning of the major substrates should be based upon understanding of protein metabolism and carbohydrate- and lipid-handling during critical illness.4 4A) In critically ill children with a functioning gastrointestinal tract, enteral nutrition (EN) should be the C preferred mode of nutrient provision, if tolerated. 4B) A variety of barriers to EN exist in the pediatric intensive care unit (PICU) Clinicians must identify and D prevent avoidable interruptions to EN in critically ill children. 4C) There are insufficient data to recommend the appropriate site (gastric vs post-pyloric/transpyloric) for enteral C feeding in critically ill children. Post-pyloric or transpyloric feeding may improve caloric intake when compared to gastric feeds. Post-pyloric feeding may be considered in children at high risk of aspiration or those who have failed a trial of gastric feeding.5 Based on the available pediatric data, the routine use of immunonutrition or immune-enhancing diets/nutrients D in critically ill children is not recommended.6 A specialized nutrition support team in the PICU and aggressive feeding protocols may enhance the overall E delivery of nutrition, with shorter time to goal nutrition, increased delivery of EN, and decreased use of parenteral nutrition. The affect of these strategies on patient outcomes has not been demonstrated.and morbidity.4,5 Children admitted to the PICU are fur- and resting energy expenditure (REE). Albumin, whichther at risk of longstanding altered nutrition status and has a large pool and much longer half-life (14-20 days), isanthropometric changes that may be associated with mor- not indicative of the immediate nutrition status. Indepen­bidity.13 Hulst et al observed a correlation between energy dently of nutrition status, serum albumin concentrationsdeficits and deterioration in anthropometric parameters may be affected by albumin infusion, dehydration, sepsis,such as mid-arm circumference and weight in a mixed trauma, and liver disease. Thus, its reliability as a markerpopulation of critically ill children.13 These anthropo- of visceral protein status is questionable. Prealbuminmetric abnormalities accrued during the PICU admis­ (also known as transthyretin or thyroxine-binding preal-sion returned to normal by 6 months after discharge.1 bumin) is a stable circulating glycoprotein synthesized inUsing reproducible anthropometric measures, Leite et al the liver. It binds with retinol binding-protein and isreported a 65% prevalence of malnutrition on admission involved in the transport of thyroxine as well as retinol.with increased mortality in this group.5 On follow up, a Prealbumin, so named by its proximity to albumin on ansignificant portion of these children had further deterio- electrophoretic strip, has a half-life of 24-48 hours.ration in nutrition status (Table 3). Nutrition assessment Prealbumin serum concentration is diminished in liverof children during the course of critical illness is desirable disease and may be falsely elevated in renal failure.and can be quantitatively assessed by routine anthropo- Prealbumin is readily measured in most hospitals and is ametric measurements. Routine monitoring of weight is a good marker for the visceral protein pool.14,15 Visceralvaluable index of nutrition status in critically ill children. proteins such as albumin and prealbumin do not accu-However, weight changes and other anthropometric rately reflect nutrition status and response to nutritionmeasurements during the PICU admission should be intervention during inflammation. In children with burninterpreted in the context of fluid therapy, other causes injury, serum acute-phase protein levels rise within 12–24of volume overload, and diuresis. Nutrition assessment hours of the stress, because of hepatic reprioritizationcan also be achieved by measuring the nitrogen balance of protein synthesis in response to injury.16 The rise is Downloaded from at International Islamic University Malaysia on February 15, 2012
  5. 5. A.S.P.E.N. Clinical Guidelines / Mehta et al   263proportional to the severity of injury. Many hospitals are Rationalecapable of measuring C-reactive protein (CRP) as an Acute injury markedly alters energy needs. Acute injuryindex of the acute-phase response. When measured seri- induces a catabolic response that is proportional to theally (once a day during the acute response period), serum magnitude, nature, and duration of the injury. Increa­prealbumin and CRP are inversely related (ie, serum pre- sed serum counter-regulatory hormone concentrationsalbumin concentrations decrease and CRP concentra- induce insulin and growth hormone resistance, resultingtions increase with the magnitude proportional to injury in the catabolism of endogenous stores of protein, carbo-severity and then return to normal as the acute injury hydrate, and fat to provide essential substrate interme-response resolves). In infants after surgery, decreases in diates and energy necessary to support the ongoingserum CRP values to levels < 2 mg/dL have been associ- metabolic stress response.19 In mechanically ventilatedated with the return of anabolic metabolism and are fol- children in the PICU, a wide range of metabolic stateslowed by increases in serum prealbumin levels.17 has been reported with an average early tendency towards hypermetabolism.20 Children with severe burn Future Research injury demonstrate extreme hypermetabolism in the early stages of injury whereby standard equations have beenStandard anthropometric measurements may be inaccu- shown to underestimate the measured REE.21 Failure torate in critically ill children with fluid shifts, edema, and provide adequate energy during this phase may result inascites. The prevalence of malnutrition in this group of loss of critical lean body mass and may worsen existingpatients and the dynamic effects of critical illness on malnutrition. Stress or activity correction factors havenutrition status require the ability to accurately measure been traditionally factored into basal energy requirementbody composition in hospitalized children. Body composi- estimates to adjust for the nature of illness, its severitytion measurement in children admitted to the PICU has and the activity level of hospi­ alized subjects.22,23 On the tbeen limited due to the absence of reliable bedside tech- other hand, critically ill children who are sedated andniques while existing measurement techniques such as mechanically ventilated may have significant reduction inthe dual energy X-ray absorptiometry (DEXA) scan are true energy expenditure, due to multiple factors includingimpractical in this cohort. Future research related to vali- decreased activity, decreased insensible fluid losses anddation of simple, noninvasive bedside body composition transient absence of growth during the acute illness.8measurement techniques is desirable and will allow moni- These patients may be at a risk of overfeeding when esti-toring of relevant parameters such as lean body mass, total mates of energy requirements are based on age-appropri-body water, and fat mass in critically ill children. Further­ ate equations developed for healthy children and especiallymore, long-term follow up studies in survivors of critical if stress factors are incorporated. The application of aillness will provide a better idea of the toll of a PICU uniform stress correction factor for broad groups ofcourse on nutrition status of children. For the purpose of patients in the ICU is simplistic, likely to be inaccuratesuch long-term follow up, qualitative markers of lean body and may increase the risk of overfeeding. IC testing maymass integrity and function or indicators of return to base- be considered before incorporating stress factor correc-line activity are examples of outcome variables relevant to tion to energy estimates in critically ill children. Therefore,nutrition in children surviving critical illness. the application of correction factors for activity, insensi- ble fluid loss and the energy or caloric allotment for growth, which is substantial in infancy, must be2. Energy Requirement in the Critically Ill Child reviewed. 2A) Energy expenditure should be assessed To account for dynamic alterations in energy meta­ throughout the course of illness to determine bolism during the critical illness course, REE values the energy needs of critically ill children. remain the only true guide for energy intake. It is likely Estimates of energy expenditure using avail- that resource constraints and lack of available expertise able standard equations are often unreliable. restricts the regular use of IC in the PICU. Estimating Grade D energy expenditure needs based on standard equations 2B) In a subgroup of patients with suspected has been shown to be inaccurate and can significantly metabolic alterations or malnutrition, accu- underestimate or overestimate the REE in critically ill rate measurement of energy expenditure using children (see Table 4). This exposes the critically ill indirect calorimetry (IC) is desirable. If IC is child to potential underfeeding or overfeeding during not feasible or available, initial energy provi- the ICU stay, with significant morbidity associated with sion may be based on published formulas or each scenario. While the problems with underfeeding nomograms. Attention to imbalance between have been well documented, overfeeding too has dele­ energy intake and expenditure will help to terious consequences.24,25 It increases ventilatory work prevent overfeeding and underfeeding in this by increasing carbon dioxide production and can poten­ population. Grade E tially prolong the need for mechanical ventilation.26 Downloaded from at International Islamic University Malaysia on February 15, 2012
  6. 6. 264   Journal of Parenteral and Enteral Nutrition / Vol. 33, No. 3, May/June 2009 Table 3 Anthropometric Changes During Pediatric Critical Illness and Role of Nutrition AssessmentStudy PopulationYear Intervention Sample Clinical OutcomeGrade Outcome Measured Size Results CommentsHulst Children in a multidisciplinary N = 261 Mean Energy deficits (over a Mixed population.   et al1,13 ICU maximum of 14 days) were Negative energy and protein2004 a) Total actual intakes of calories      27 kcal/kg—Preterm neonates balance in this populationLevel III and protein were recorded.      20 kcal/kg—Term neonates correlated with decreasing Balance was calculated by sub-      12 kcal/kg—Older children anthropometric parameters. tracting actual intake from RDA, Mean Protein deficits: A 14-day period of monitoring over a maximum of 14 days.      0.6 g/kg/day—Preterm may not be adequate for Relations between balance and      0.3 g/kg/day—Term Newborns anthropometric changes. clinical factors and change in      0.2 g/kg/day—Children Energy balance was anthropometry Cumulative deficits—related to calculated from estimates of b) Patients were also followed up decrease in weight and arm RDA and not measured by to 6 months for anthropometric circumference SD-scores. indirect calorimetry. parameters. Negative correlation with age, At 6 months follow up, almost length of stay in the ICU and all children had recovered duration of mechanical their nutrition status. ventilator support.Hulst et al18 Children in a multidisciplinary N = 105 Prevalence of hypomagnesemia, Except for uremia, no2006 PICU hypertriglyceridemia, uremia significant association wasLevel III Serum urea, albumin, triglycerides and hypoalbuminemia were found between and magnesium were measured 20%, 25%, 30% and 52%, abnormalities in in 105 children (age, 7 days to respectively, with no significant biochemical parameters and 16 years) within the first 24 associations between the changes in SD scores of hours after admission. different disorders. anthropometric Association with anthropometric measurements. outcomes parametersLeite et al5 PICU N = 46 65% of the patients presented with A significant number of1993 Anthropometry at admission and indices of malnutrition. Of patients are nutritionally-at-Level III follow-up these, chronic malnutrition was risk at the time of hospital predominant. admission, and there is an Mortality was higher in association between malnourished individuals (20% nutrition status and hospital vs 12.5%). course. 36% of patients showed a decrease The anthropometric nutrition in weight-for-height on follow evaluation is a simple, up. reproducible and objective tool for nutrition assessment of the critically ill child.ICU, intensive care unit; PICU, pediatric intensive care unit; RDA, recommended daily allowance; SD, standard deviation.Overfeeding may also impair liver function by inducing feeding in a select group of chronically ill children at highsteatosis and cholestasis, and increase the risk of infection risk of obesity is currently sporadic. In general, the energysecondary to hyperglycemia. Hyperglycemia associated goals should be assessed and reviewed regularly inwith caloric overfeeding has been associated with critically ill children.prolonged mechanical ventilator requirement and PICU Table 4 summarizes studies examining the performanceLOS.27 The use of the respiratory quotient (RQ) as a of estimated energy needs in relation to measured REE inmeasure of substrate use in individual children cannot be critically ill children requiring mechanical ventilatorrecommended. However, a combination of acute phase support. In general, these small sized, prospective orproteins (CRP) and RQ may reflect transi­ ion from the t retrospective cohort studies demonstrate the variability ofcatabolic hypermetabolic to the anabolic state. There are the metabolic state and the uniform failure of estimatedno data in general pediatric populations for the role of energy needs in accurately predicting the measured REEhypocaloric feeding. The application of hypocaloric in critically ill children. In the absence of REE, some Downloaded from at International Islamic University Malaysia on February 15, 2012
  7. 7. A.S.P.E.N. Clinical Guidelines / Mehta et al   265 Table 4 Estimated Energy Expenditure vs Measured Resting Energy Expenditure (REE)Study PopulationYear Intervention Clinical OutcomeGrade Outcome Measures Sample Size Results CommentsDe Klerk Children needing > 24 hours of N = 18 Variability in total daily energy Single measurement et al28 mechanical ventilator support expenditure appears to accurately2002 in a PICU (40-64 kcal/kg/d) reflect total daily energyLevel III Serial measured REE CV was < 10% requirement. Respiratory Quotient (RQ) Positive energy balance in Results of this study do not Energy balance = actual energy many children (n=8) suggest the need for intake – total energy [Average RQ in this group was serial REE. expenditure 0.89] RQ was marginally affected Negative balance (n=10) by energy balance. [Average RQ in this group was 0.84]White et al29 Mechanically ventilated children N = 11 30-minute REE Small numbers. No1999 in the PICU CV was 7.2% ± 4.5% correlation examinedLevel III 24 h indirect calorimetry 30-min vs 24 h: no difference with clinical state. measurement (P < 0.69) 30-minute REE accurately 30 minute steady state REE No diurnal variation represented the 24-h compared to 24-h total energy Between-day CV = 21% ± 16% values. expenditure The authors recommend Daily CV in measured REE was serial REE measure- calculated ments based on signi- ficant between-day CV.White et al30 Mechanically ventilated children N = 100 A new equation for estimated The authors concluded2000 in a PICU (derivation) REE was derived, that there is noLevel III Clinical variables N = 25 incorporating age, weight, substitute for measured REE by indirect calorimetry (validation) temperature, days in the REE. PICU, and disease. Their derived equation performed better than existing standard equations.Derumeaux- Obese children N = 471 REE equation using fat-free Special equations may be Burel BMI z score ≥ 2 (derivation) mass was more accurate than necessary for obese et al31 Measured REE N = 211 weight-based equations children.2004 Fat free mass—obtained by (validation) Body composition is anLevel III bioelectric impedance important factor in REE. assessment Measured REE is ideal. Predicted equationsMlcak et al32 Children < 18 years with total N = 100 (40 REE was expressed in 3 Increased REE persisted2006 body surface area burn > 40%, female) different ways: actual REE in for over 12 months afterLevel III and consent to return at 6, 9, kcal per day, percent of burn injury. and 12 months for post-burn predicted REE, and actual follow up REE divided by the BMI. Measured REE vs Harris- Hypermetabolism persisted 12 Benedict equation and months after burn injury. corrected by BMI Female children exerted a REE measurements were decreased hypermetabolic repeated at 6, 9, and 12 response compared with months post-burn when the male children. patients returned for outpatient surgery. (continued) Downloaded from at International Islamic University Malaysia on February 15, 2012
  8. 8. 266   Journal of Parenteral and Enteral Nutrition / Vol. 33, No. 3, May/June 2009 Table 4 (continued)Study PopulationYear Intervention Clinical OutcomeGrade Outcome Measures Sample Size Results CommentsFramson Children admitted to a PICU. N = 44 (29 In general, equations The hypermetabolic et al33 Both spontaneously breathing males) performed well. response apparent in2007 and mechanically ventilated Mean REE for all adults was not evidentLevel III patients were included. No measurements was 821 ± in these critically children with chronic disease. 653 kcal/24 hours. ill children. The Schofield equation REE measurement within 24 estimate was 798 ± 595 Authors do not recommend hours of admission, then 48 kcal/24 h and the White the use of REE hours after the first equation estimate was 815 ± estimates from equations measurement, and finally 564 kcal/24 h (P not as a guide for caloric within 24 hours before significant). intake in critically ill discharge from the PICU. Only 45% of REE were within children. Measured REE was compared to 90%–110% of that predicted estimates from Schofield and by the Schofield equation. White equations. The White equation was inaccurate (not within 10% of REE) in 66 of 94 measurements (70%). The discrepancy was greatest (100%) in children with REE <450 kcal/24 h.Vazquez Mechanically ventilated children N = 43 (18 Measured REE = 674 ± 384 Children may be Martinez in a PICU female); 35 kcal/day. hypometabolic in the et al34 Measured REE was compared to surgical and Patients noted to be first 6 h after PICU2004 estimates from various 8 medical hypometabolic in the first 6 admission.Level III equations/formulas such as h after admission to PICU. Harris-Benedict, Caldwell- Most equations overestimated Equations overestimate Kennedy, Schofield, FAO/ measured REE in ventilated, energy expenditure. WHO, Maffeis, Fleisch, critically ill children during Authors do not recommend Kleiber, Dreyer, and Hunter the early post-injury period. equations for predicting equations. Measured and predicted energy energy expenditure in expenditure differed critically ill children. significantly (P < .05) except when the Caldwell-Kennedy and the Fleisch equations were used.BMI, body mass index; CV, coefficient of variation; FAO/WHO, Food and Agriculture Organization/World Health Organization;PICU, pediatric intensive care unit.investigators recommend that basal energy requirements Future Researchshould be provided without correction factors to avoidthe provision of calories and/or nutrition substrates in IC remains sporadically applied in critically illexcess of the energy required to maintain the metabolic children in the setting of mounting evidence of thehomeostasis of the injury response. Criteria for targeting inaccuracy of estimated basal metabolic rate usinga select group of children in the PICU for IC measurement standard equations. This could potentially subject a sub­of REE may be useful for centers with limited resources group of children in the PICU to the risk of underfeedingfor metabolic testing. Some children in the PICU are or overfeeding. In the era of resource constraints, IC maylikely to be at risk of altered metabolism or malnutrition, be applied or targeted for certain high-risk groups in thewhere estimates of energy expenditure using standard PICU. Selective application of IC may allow many unitsequations are likely to be inaccurate. If resources are to balance the need for accurate REE measurement andlimited, this subset of the population may benefit from limited resources (Appendix 1). Studies examining thetargeted IC for accurate measurement of REE to guide role of simplified IC technique, its role in optimizingenergy administration. nutrient intake, its ability to prevent overfeeding or Downloaded from at International Islamic University Malaysia on February 15, 2012
  9. 9. A.S.P.E.N. Clinical Guidelines / Mehta et al   267underfeeding in selected subjects, and the cost-benefit response. This allows for maximal physiologic adaptabilityanalyses of its application in the PICU are desirable. The at times of injury or illness. Although children with criti-effect of energy intake on outcomes needs to be examined cal illness have increases in both whole-body proteinin pediatric populations especially in those on the extremes degradation and whole-body protein synthesis, it is theof body mass index (BMI). former that predominates during the stress response. Thus, these patients manifest net negative protein and nitrogen balance characterized by skeletal muscle wast- Appendix 1 ing, weight loss, and immune dysfunction. The catabo- Children at high risk for metabolic alterations who lism of muscle protein to generate glucose andare suggested candidates for targeted measurement of inflammatory response proteins is an excellent short-termREE in the PICU include the following: adaptation, but it is ultimately limited because of the reduced protein reserves available in children and neo- • Underweight (BMI < 5th percentile for age), at risk nates. Unlike during starvation, the provision of dietary of overweight (BMI > 85th percentile for age) or carbohydrate alone is ineffective in reducing the endoge- overweight (BMI > 95th percentile for age) nous glucose production via gluconeogenesis in the • Children with > 10% weight gain or loss during metabolically stressed state.35 Therefore, without elimina- ICU stay tion of the inciting stress for catabolism (ie, the critical • Failure to consistently meet prescribed caloric goals • Failure to wean, or need to escalate respiratory illness or injury), the progressive breakdown of muscle support mass from critical organs results in loss of diaphragmatic • Need for muscle relaxants for > 7 days and intercostal muscle (leading to respiratory compro- • Neurologic trauma (traumatic, hypoxic and/or isch- mise), and to the loss of cardiac muscle. The amount of emic) with evidence of dysautonomia protein required to optimally enhance protein accretion is • Oncologic diagnoses (including children with stem higher in critically ill than in healthy children. Infants cell or bone marrow transplant) demonstrate 25% higher protein degradation after surgery • Children with thermal injury and a 100% increase in urinary nitrogen excretion with • Children requiring mechanical ventilator support bacterial sepsis.36,37 The provision of dietary protein suf- for > 7 days ficient to optimize protein synthesis, facilitate wound • Children suspected to be severely hypermetabolic healing and the inflammatory response, and preserve (status epilepticus, hyperthermia, systemic inflam- matory response syndrome, dysautonomic storms, skeletal muscle protein mass is the most important nutri- etc) or hypometabolic (hypothermia, hypothyroid- tion inter­ ention in critically ill children. The quantities v ism, pentobarbital or midazolam coma, etc.) of protein recommended for critically ill neonates and • Any patient with ICU LOS > 4 weeks may benefit children are based on limited data. Certain severely from IC to assess adequacy of nutrient intake. stressed states, such as significant burn injury, may require additional protein supplementation to meet meta- bolic demands. Excessive protein administration should3. Macronutrient Intake During Critical Illness be avoided as toxicity has been documented, particularly There are insufficient data to make evidence- in children with marginal renal and hepatic function. based recommendations for macronutrient Studies using high protein allotments of 4–6 g/kg/day intake in critically ill children. After determi- have been associated with adverse effects such as azotemia, nation of energy needs for the critically ill metabolic acidosis, and neurodevelopmental abnormali- child, the rational partitioning of the major ties.38 A similar evaluation of the effects of high protein substrates should be based upon basic under- administration using newer formulas is desirable. Although standing of protein metabolism and carbohy- the precise amino acid composition to best increase drate- and lipid-handling during critical illness. whole-body protein balance has yet to be fully deter- Grade E mined, stable isotope techniques now exist to study this issue. Estimated protein requirements for injured children of various age groups are as follows: 0–2 years, Rationale 2–3 g/kg/day; 2–13 years, 1.5–2 g/kg/day; and 13–18Critical illness and recovery from trauma or surgery are years, 1.5 g/kg/day.characterized by increased protein catabolism and turn- Once protein needs have been met, safe caloric provi­over. An advantage of high protein turnover is that a con- sions using carbohydrate and lipid energy sources havetinuous flow of amino acids is available for synthesis of similar beneficial effects on net protein synthesis andnew proteins. Specifically, this process involves a redistri- overall protein balance in critically ill patients. Glucose isbution of amino acids from skeletal muscle to the liver, the primary energy used by the brain, erythrocyte, andwound, and other tissues involved in the inflammatory renal medulla and is useful in the repair of injured tissue. Downloaded from at International Islamic University Malaysia on February 15, 2012
  10. 10. 268   Journal of Parenteral and Enteral Nutrition / Vol. 33, No. 3, May/June 2009Glycogen stores are limited and quickly depleted in illness generally restricted to a maximum of 30%–40% of totalor injury, resulting in the need for gluconeogenesis. In calories, although this practice has not been validated byinjured and septic adults, a 3-fold increase in glucose clinical trials.turnover and oxidation has been demonstrated as well asan elevation in gluconeogenesis. A significant feature of 4. Route of Nutrient Intake (Enteral Nutrition)the metabolic stress response is that the provision ofdietary glucose does not halt gluconeogenesis. Conse­ 4A) In critically ill children with a functioningquently, the catabolism of muscle protein to produce gastrointestinal tract, enteral nutrition (EN)glucose continues unabated, and attempts to provide should be the preferred mode of nutrient pro-large carbohydrate intake in critically ill patients have vision, if tolerated. Grade Cbeen abandoned. 4B) A variety of barriers to EN exist in the PICU. The Surviving Sepsis Campaign has recommended Clinicians must identify and prevent avoidabletight glucose control in critically ill adults based on interruptions to EN in critically ill children.results of a single trial that showed decreased mortality in Grade Dcritically ill adults randomized to this strategy. Subsequent 4C) There are insufficient data to recommendstudies examining the role of strict glycemic control in the appropriate site (gastric vs post-pyloric/adults have yielded conflicting results and the incidence transpyloric) for enteral feeding in critically illof hypoglycemia in these studies is concerning.39 children. Post-pyloric or transpyloric feedsHyperglycemia is prevalent in critically ill children and may improve caloric intake when compared tohas been associated with poor outcomes in retrospective gastric feeds. Post-pyloric feeding may be con-studies.27,40,41 The etiology of hyperglycemia during the sidered in children at high risk of aspiration orstress response is multifactorial. Despite the prevalence those who have failed a trial of gastric feeding.of hyperglycemia in the pediatric intensive care population, Grade Cno data exist currently evaluating the effects of tightglycemic control in the pediatric age group. Both Rationalehypoglycemia and glucose variability also are associatedwith increased LOS and mortality, and hence are Following the determination of energy expenditure andundesirable in the critically ill child.42 In the absence of requirement in the critically ill child, the next challenge is todefinitive data, aggressive glycemic control cannot be select the appropriate route for delivery of nutrients. In therecommended as yet in the critically ill child. critically ill child with a functioning gastrointestinal tract, Lipid turnover is generally accelerated by critical the enteral route is preferable to parenteral nutrition (PN).illness, surgery, and trauma.43 Recently, it has been EN has been shown to be more cost-effective without theshown that critically ill children do, indeed, have a added risk of nosocomial infection inherent with PN.47,48higher rate of fat oxidation.44 Thus, this suggests that However, the optimal route of nutrient delivery has not beenfatty acids are, in fact, the prime source of energy in systematically studied in children and there is no RCT com-metabolically stressed children. Because of the increased paring the effects of EN vs PN. Current practice in manydemand for lipid use in critical illness coupled with the centers includes the initiation of gastric or post-pyloriclimited fat stores in the pediatric patient, critically ill enteral feeding within 48-72 hours after admission. PN ischildren are susceptible to the evolution of biochemically being used to supplement or replace EN in those patientsdetected essential fatty acid deficiency if administered a where EN alone is unable to meet the nutrition goal.fat-free diet.45 Clinically, this syndrome presents as In children fed with EN, there are insufficient data todermatitis, alopecia, thrombo­ ytopenia, and increased c make recommendations regarding the site of enteralsusceptibility to bacterial infec­ ion. To avoid essential t feeding (gastric vs post-pyloric). Meert et al examined thefatty acid deficiency in critically ill or injured infants, role of small bowel feeding in 74 critically ill children,the allotment of linoleic and linolenic acid is randomized to receive either gastric or post-pyloricrecommended at concentrations of 4.5% and 0.5% of nutrition.49 The study was not powered to detecttotal calories, respectively. The provision of commercially differences in mortality. EN was interrupted in a largeavailable intravenous fat emulsions (IVFE) to parenterally number of subjects in this study and caloric goals werefed critically ill children reduces the risk of essential met in a small percentage of the population studied. Thisfatty acid deficiency, results in improved protein use, unblinded RCT did not show difference in microaspiration,and does not significantly increase CO2 production or enteral access device displacement, and feed intolerancemetabolic rate.46 Most centers, therefore, start IVFE between the gastric or post-pyloric fed groups. A highersupplementation in critically ill children at 1 g/kg/day percentage of subjects in the small bowel group achievedand advance over a period of days to 2-4 g/kg/day, with their daily caloric goal compared to the gastric fed group.monitoring of triglyceride levels. IVFE administration is Sanchez et al report better tolerance in critically ill Downloaded from at International Islamic University Malaysia on February 15, 2012
  11. 11. A.S.P.E.N. Clinical Guidelines / Mehta et al   269children receiving early (< 24 hours after PICU admission) population. Consistently underachieved EN goals arevs late (started after 24 hrs) post-pyloric nutrition.50 Of thought to be one of the reasons for the absence ofthe 526 children in their cohort who were deemed to have beneficial effect in multiple studies and meta-analysis ofintolerance to EN, 202 received early post-pyloric the efficacy of immunonutrition in preventing infection.56nutrition and had decreased incidence of abdominal Awareness of these factors hindering the achievement ofdistension. Despite evidence to suggest that it is reasonably EN goals is essential in order to address preventabletolerated, the routine use of postpyloric feeding in the interruptions in enteral feeding in critically ill children.critically ill child cannot be recommended. It may be There is not enough evidence to recommend the use ofprudent to consider this option in patients who do not prokinetic medications or motility agents (for ENtolerate gastric feeding or those who are at a high risk of intolerance or to facilitate enteral access deviceaspiration. Postpyloric or transpyloric feeding may be placement), prebiotics, probiotics, or synbiotics inlimited by the ability to obtain small bowel access, and critically ill children.the expertise and resources in individual PICUs are likelyto be variable. A standardized approach to optimizing Future Researchbenefits and minimizing risks with EN delivery will helpclinicians identify patients who would benefit from small Future studies may be directed at examining methods tobowel feeding. ensure optimal prescription and delivery of nutrient Despite the absence of sound evidence to support the intake at the bedside, identifying and preventing commonsuperiority of one route of feeding over the other, the reasons for avoidable interruptions in nutrient intake,enteral route has been successfully used for nutrition selection of children at risk of aspiration in the PICU,support of the critically ill child.51-53 In another unblinded and the role of EN (gastric vs postpyloric feeds) in thisRCT, Horn et al randomized 45 children admitted to subgroup. The advantages of EN in terms of its role in gutthe PICU to receive gastric tube feeding either continu­ immunity, prevention of PN related complications andously or intermittently every 2 hours. The main outcome the cost benefit analysis when compared to PN requiremeasure examined in this study was tolerance of enteral further evaluation.feedings. The small sample size and the short observationperiod of < 66 hours makes any meaningful interpretation 5. Immunonutrition in the PICUdifficult. However, the number of daily stools, diarrhealepisodes, or vomiting episodes was similar between the 2 Based on the available pediatric data, the routinegroups. Intolerance to enteral feedings may limit intake use of immunonutrition or immune-enhancingand supplementation with PN may be required. Prospective diets/nutrients in critically ill children is notcohort studies and retrospective chart reviews have recommended. Grade Dreported the inability to achieve daily caloric goal incritically ill children.54,55 The most common reasons for Rationalesuboptimal enteral nutrient delivery in these studies arefluid restriction, interruptions to EN for procedures, and The use of specific nutrients aimed at modulating theEN intolerance due to hemodynamic instability. The inflammatory or immune response has been reported forpercent of estimated energy expenditure actually several years. Despite several RCTs employing immunonu­administered to these subjects was remarkably low. In a trition in critically ill patients, a positive treatment effectstudy examining the endocrine and metabolic response of of immunonutrition or the use of immune-enhancingchildren with meningococcal sepsis, goal nutrition was diets (IED) has not been demonstrated. These studiesachieved in only 25% of the cases.19 Similar observations are flawed by their poor methodology and small samplehave been made in a group of 95 children in a PICU size. The studies were conducted using a variety of nutri-where patients received a median of 58.8% (range ents in combination that were administered to heteroge-0%-277%) of their estimated energy requirements. In this neous patient populations. The studies do not allowreview, EN was interrupted on 264 occasions for clinical meaningful interpretation of the safety or efficacy of indi-procedures. In another review of nutrition intake in 42 vidual nutrients and fail to detect significant differencespatients in a tertiary-level PICU over 458 ICU days, in relevant clinical outcomes. Arginine, glutamine, amin-actual energy intake was compared with estimated energy opeptides, w-3 fatty acids and antioxidants are some ofrequirement.55 Only 50% of patients were reported to the nutrients studied for their immune modulation effects.have received full estimated energy requirements after a Systematic reviews of immunonutrition studies in adultsmedian of 7 days in the ICU. Protocols for feeding use of have cautioned against the use of arginine and othertranspyloric feeding tubes and changing from bolus to nutrients due to potential for harm in septic and criticallycontinuous EN during brief periods of intolerance are ill patients.59 Fish oils, borage oils, and antioxidantsstrategies to achieve estimated energy goals in this may have a role in patients with acute respiratory distress Downloaded from at International Islamic University Malaysia on February 15, 2012
  12. 12. 270   Journal of Parenteral and Enteral Nutrition / Vol. 33, No. 3, May/June 2009 Table 5 Clinical Outcomes Associated With Enteral FeedingStudy PopulationYear Intervention Sample Clinical OutcomeGrade Outcome Measures Size Results CommentsMeert et al49 Critically ill children (<18yrs) in N = 74 Daily caloric goal achieved was Small number of patients2004 a PICU significantly lower (p < 0.01) studied.Level II RCT comparing gastric vs small in gastric group (30 ± 23%) vs Difficult to blind such a bowel continuous tube feeds small bowel group (47 ± 22%). study at the bedside (due Percentage of caloric goals Proportion of patients with to need for radiographic achieved, pepsin positive microaspiration, tube confirmation of tracheal secretions, and feed displacement, and EN placement of tip of tolerance (vomiting, diarrhea intolerance were similar enteral access device). or abdominal distension) were between the 2 groups. Fairly high number of the main outcome variables. subjects in the study experienced EN interruptions and percent of caloric goals met was low in both groups.Horn et al57 Children < 18y in a PICU with N = 45 Number of stools/d (1.5 vs 1.6), Small number of patients2003 EN mean episodes of diarrhea/d studied.Level II RCT comparing gastric EN (0.32 vs 0.64), mean number The duration of study was administered either of vomiting episodes/person too small to detect continuously or every 2 h (0.64 vs 0.22) were similar meaningful differences— Tolerance—number and type of between the continuous and median 64.5 and 66 stools, diarrhea and vomiting intermittent gastric fed group. hours for the 2 groups.De Oliveira Children in a PICU with N = 55 EN started on day 3 ± 1 and Energy requirements were Iglesias EN ≥ 2 d maintained for 6 ± 3 d. 71% estimates and not et al58 Required (estimates) vs received the required calories measured.2007 prescribed vs. delivered calories (38% of which reached caloric Barriers to EN remain aLevel III were recorded. goal by d 5 of PICU significant challenge to Variables associated with not admission). Daily average ensuring the delivery of achieving caloric goals were caloric intake was 60% of prescribed calories to recorded. required and 85% of patients in the PICU. prescribed. Barriers to EN were procedures, clinical instability, use of inotropic agents, enteral access device displacement, postoperative fasting, and feeding intolerance (abdominal distension, vomiting, and diarrhea).Sanchez Children over 10 years of age N = 526 Clinical characteristics, nutrients Retrospective cohort et al50 admitted to a PICU were (202 were delivered and incidence of Children tolerate early2007 eligible for transpyloric tube fed early) diarrhea were similar in both postpyloric feeds.Level III placement if they were deemed groups. to not tolerate enteral feedings Abdominal distension was less within 24-48 hrs after frequent in early EN group admission. (3.5%) vs late (7.8%); P < .05. Tolerance of feeds was compared between patients started on transpyloric feeds early (< 24 hours after admission) vs those started late (> 24 hrs) Calories delivered, duration of nutrition, abdominal distension and diarrhea were recorded. (continued) Downloaded from at International Islamic University Malaysia on February 15, 2012
  13. 13. A.S.P.E.N. Clinical Guidelines / Mehta et al   271 Table 5 (continued)Study PopulationYear Intervention Sample Clinical OutcomeGrade Outcome Measures Size Results CommentsRogers Children admitted to a PICU N = 42 Patients received 37.7% (median) of PICU patients, especially et al55 with length of stay >3 d (18 Cardiac their EER. cardiac surgical patients,2003 and EN intake surgical) Cardiac surgical patients had lower do not receive their energyLevel III The proportion of patients caloric intake than others. goals. Fluid restriction in reaching estimated energy Only 52% of the pts received full both groups is the major requirement goals was energy intake goal at any time reason for inability to meet recorded. during the PICU stay. Significant energy goals. weight decrease was noted in Fasting for procedures and cardiac surgical patients. EN interruptions due to Major barrier to caloric intake—fluid intolerance were other restriction barriers to nutrition in Minor barriers—feed interruption this study. for procedures and intoleranceTaylor Children admitted to PICU N = 95 59% were fed within 24 hours of Poor % energy intake goal et al54 with length of stay ≥3 d admission. achieved with EN.2003 Information on nutrition EN was administered 54% of time. Procedure-related feedingLevel III delivery was recorded. 10% received PN; 9.5% did not interruptions were receive any nutrition support. significant. PICU pts received 58.8%, median The study highlights the (range 0–277%) of their energy need for a proactive intake goal. approach to bridge the Energy intake was greater when gap between desirable supplemented by PN. and achieved nutrient EN was interrupted 264 times intake. (mainly for procedures). For up to 75% of study time, children had abnormal bowel patterns. 79% were constipated for 3-21 days. And 43% had diarrhea of unknown etiology.EER, energy efficiency ratio; EN, enteral nutrition; PICU, pediatric intensive care unit; RCT, randomized controlled trial.syndrome (ARDS). Glutamine may have beneficial effects Another small pilot RCT reported improved outcomesin adults with burn injury and trauma. in children fed with a glutamine-enriched formulation, The role of immune-enhancing EN in children during although the numbers are too small for meaningfulcritical illness has not been extensively studied. Briassoulis conclusions.60 The use of a specialized adult immuneet al reported their results of a blinded RCT in children modulating enteral formula in pediatric burn victims hasadmitted to the PICU with expected LOS and need for been associated with improvement in oxygenation andmechanical ventilation of ≥ 5 d.22 EN was started in these pulmonary compliance in a retrospective review.61patients within 12 h of admission to PICU. Patients wererandomized to receive either a formulation containing Future Researchglutamine, arginine, w-3 fatty acids, and antioxidants orstandard age-appropriate formulation. Protocolized increase Future pediatric studies in this field must focus on exam-in EN ensured that goal feeds were reached by day 4. The ining the effects of single (vs combination of) nutrients,study did not show any outcome differences in the 25 in large (multicenter) trials, on homogeneous PICUchildren in each arm, although authors report a trend populations designed to detect differences in importanttoward a decrease in nosocomial infection rates and outcome measures. This approach will ensure that resultspositive gastric aspirate culture rates in the treatment arm. allow meaningful inferences to be made about soundThe immunologically active formula used in this study was hypotheses on single immune modulating nutrients andnot specifically tailored for children and transient diarrhea will prevent the current absence of strong conclusionswas noted in children receiving this formula, which had despite a large amount of investment in this area ofa higher osmolarity compared to the control formula. research in adult ICU populations. Downloaded from at International Islamic University Malaysia on February 15, 2012