Guidelines for the Provision and Assessment of Nutrition Support Therapy in the Adult Critically Ill Patient:

Practice guidelines are not intended as absolute requirements. The use of these practice guidelines does not in any way project or guarantee any specific benefit in outcome or survival. The judgment of the healthcare professional based on individual circumstances of the patient must always take precedence over the recommendations in these guidelines. The guidelines offer basic recommendations that are supported by review and analysis of the pertinent available current literature, by other national and international guidelines, and by the blend of expert opinion and clinical practicality. The “intensive care unit” (ICU) or “critically ill” patient is not a homogeneous population. Many of the studies on which the guidelines are based are limited by sample size, patient heterogeneity, variability in definition of disease state and severity of illness, lack of baseline nutrition status, and lack of statistical power for analysis. Whenever possible, these factors are taken into account and the grade of statement will reflect the power of the data. One of the major methodological problems with any guideline is defining the exact population to be included. These guidelines may be subject to periodic review and revision based on new peer-reviewed critical care nutrition literature and practice. These guidelines are intended for the adult medical and surgical critically ill patient populations expected to require an ICU stay of > 2 or 3 days and are not intended for those patients in the ICU for temporary monitoring or those who have minimal metabolic or traumatic stress. These guidelines are based on populations, but like any other therapeutic treatment in an ICU patient, nutrition requirements and techniques of access should be tailored to the individual patient. The intended use of these guidelines is for all individuals involved in the nutrition therapy of the critically ill, primarily physicians, nurses, dietitians, pharmacists, and respiratory and physical therapists where indicated. A list of guideline recommendations was compiled by the experts on the Guidelines Committee for the 2 societies, each of which represented clinically applicable definitive statements of care or specific action statements. Prospective randomized controlled trials were used as the primary source to support guideline statements, with each study being evaluated and given a level of evidence. The overall grade for the recommendation was based on the number and level of investigative studies referable to that guideline. Large studies warranting level I evidence were defined as those with ≥100 patients or those which fulfilled endpoint criteria predetermined by power analysis. The level of evidence for uncontrolled studies was determined by whether they included contemporaneous controls (level III), historical controls (level IV), or no controls (level V, equal to expert opinion). See Table 1 . 1 Review papers and consensus statements were considered expert opinion and were designated the appropriate level of evidence. Meta-analyses were used to organize the information and to draw conclusions about an overall treatment effect from multiple studies on a particular subject. The grade of recommendation, however, was based on the level of evidence of the individual studies. An A or B grade recommendation required at least 1 or 2 large positive randomized trials supporting the claim, while a C grade recommendation required only 1 small supportive randomized investigation. The rationale for each guideline statement was used to clarify certain points from the studies, to identify controversies, and to provide clarity in the derivation of the final recommendation. Significant controversies in interpretation of the literature were resolved by consensus of opinion of the committee members, which in some cases led to a downgrade of the recommendation. Following an extensive review process by external reviewers, the final guideline manuscript was reviewed and approved by A.S.P.E.N. Board of Directors and SCCM's Board of Regents and Council. The significance of nutrition in the hospital setting cannot be overstated. This significance is particularly noted in the ICU. Critical illness is typically associated with a catabolic stress state in which patients commonly demonstrate a systemic inflammatory response. This response is coupled with complications of increased infectious morbidity, multi-organ dysfunction, prolonged hospitalization, and disproportionate mortality. Over the past 3 decades, the understanding of the molecular and biological effects of nutrients in maintaining homeostasis in the critically ill population has made exponential advances. Traditionally, nutrition support in the critically ill population was regarded as adjunctive care designed to provide exogenous fuels to support the patient during the stress response. This support had 3 main objectives: to preserve lean body mass, to maintain immune function, and to avert metabolic complications. Recently these goals have become more focused on nutrition therapy, specifically attempting to attenuate the metabolic response to stress, to prevent oxidative cellular injury, and to favorably modulate the immune response. Nutritional modulation of the stress response to critical illness includes early enteral nutrition, appropriate macro- and micronutrient delivery, and meticulous glycemic control. Delivering early nutrition support therapy, primarily using the enteral route, is seen as a proactive therapeutic strategy that may reduce disease severity, diminish complications, decrease length of stay in the ICU, and favorably impact patient outcome. A1. Traditional nutrition assessment tools (albumin, prealbumin, and anthropometry) are not validated in critical care. Before initiation of feedings, assessment should include evaluation of weight loss and previous nutrient intake prior to admission, level of disease severity, comorbid conditions, and function of the gastrointestinal (GI) tract. (Grade: E) Rationale. In the critical care setting, the traditional protein markers (albumin, prealbumin, transferrin, retinol binding protein) are a reflection of the acute phase response (increases in vascular permeability and reprioritization of hepatic protein synthesis) and do not accurately represent nutrition status in the ICU setting. Anthropometrics are not reliable in assessment of nutrition status or adequacy of nutrition therapy.2,3 A2. Nutrition support therapy in the form of enteral nutrition (EN) should be initiated in the critically ill patient who is unable to maintain volitional intake. (Grade: C) Rationale. EN supports the functional integrity of the gut by maintaining tight junctions between the intraepithelial cells, stimulating blood flow, and inducing the release of trophic endogenous agents (such as cholecystokinin, gastrin, bombesin, and bile salts). EN maintains structural integrity by maintaining villous height and supporting the mass of secretory IgA-producing immunocytes which comprise the gut-associated lymphoid tissue (GALT) and in turn contribute to mucosal-associated lymphoid tissue (MALT) at distant sites such as the lungs, liver, and kidneys.4-7 Adverse change in gut permeability from loss of functional integrity is a dynamic phenomenon which is time-dependent (channels opening within hours of the major insult or injury). The consequences of the permeability changes include increased bacterial challenge (engagement of GALT with enteric organisms), risk for systemic infection, and greater likelihood of multi-organ dysfunction syndrome (MODS).4,5 As disease severity worsens, increases in gut permeability are amplified and the enteral route of feeding is more likely to favorably impact outcome parameters of infection, organ failure, and hospital length of stay (compared to the parenteral route).8 The specific reasons for providing early EN are to maintain gut integrity, modulate stress and the systemic immune response, and attenuate disease severity.6,8,9 Additional endpoints of EN therapy include use of the gut as a conduit for the delivery of immune-modulating agents and use of enteral formulations as an effective means for stress ulcer prophylaxis. Nutrition support therapy (also called “specialized” or“ artificial” nutrition therapy) refers to the provision of enteral tube feeding or parenteral nutrition. “Standard therapy” refers to a patient's own volitional intake without provision of specialized nutrition support therapy. The importance of promoting gut integrity with regard to patient outcome is being strengthened by clinical trials comparing critically ill patients fed by EN to those receiving standard (STD) therapy. In a recent meta-analysis10 in elective gastrointestinal surgery and surgical critical care, patients undergoing a major operation who were given early postoperative EN experienced significant reductions in infection (relative risk [RR] = 0.72; 95% confidence interval [CI] 0.54-0.98; P = .03), hospital length of stay (mean 0.84 days; range 0.36-1.33 days; P = .001), and a trend toward reduced anastomotic dehiscence (RR = 0.53; 95% CI 0.26-1.08; P = .08), when compared to similar patients receiving no nutrition support therapy.10-16 In a meta-analysis17 of patients undergoing surgery for complications of severe acute pancreatitis, those placed on EN 1 day postop showed a trend toward reduced mortality compared to controls randomized to STD therapy (RR = 0.26; 95% CI 0.06-1.09; P = .06).17-19 See Table 2 . 11-16,18,19 A3. EN is the preferred route of feeding over parenteral nutrition (PN) for the critically ill patient who requires nutrition support therapy. (Grade: B) Rationale. In the majority of critically ill patients, it is practical and safe to utilize EN instead of PN. The beneficial effects of EN when compared to PN are well documented in numerous prospective randomized controlled trials involving a variety of patient populations in critical illness, including trauma, burns, head injury, major surgery, and acute pancreatitis.8,20-22 While few studies have shown a differential effect on mortality, the most consistent outcome effect from EN is a reduction in infectious morbidity (generally pneumonia and central line infections in most patient populations, and specifically abdominal abscess in trauma patients).20 In many studies, further benefits are seen from significant reductions in hospital length of stay,21 cost of nutrition therapy,21 and even return of cognitive function (in head injury patients).23 All 6 meta-analyses that compared EN to PN showed significant reductions in infectious morbidity with use of EN.21,24-28 Noninfective complications (risk difference = 4.9; 95% CI 0.3-9.5; P =.04) and reduced hospital length of stay (weighted mean difference [WMD] = 1.20 days; 95% CI 0.38-2.03; P = .004) were seen with use of EN compared to PN in 1 metaanalysis by Peter et al.28 Five of the meta-analyses showed no difference in mortality between the 2 routes of nutrition support therapy.21,24,26-28 One meta-analysis by Simpson and Doig25 showed a significantly lower mortality (RR = 0.51; 95% CI 0.27-0.97; P =.04) despite a significantly higher incidence of infectious complications (RR = 1.66; 95% CI 1.09-2.51; P =.02) with use of PN compared to EN.25 See Table 3 . 8,20,22,29-61 A4. Enteral feeding should be started early within the first 24-48 hours following admission. (Grade: C) The feedings should be advanced toward goal over the next 48-72 hours. (Grade: E) Rationale. Attaining access and initiating EN should be considered as soon as fluid resuscitation is completed and the patient is hemodynamically stable. A “window of opportunity” exists in the first 24-72 hours following admission or the onset of a hypermetabolic insult. Feedings started within this time frame (compared to feedings started after 72 hours) are associated with less gut permeability, diminished activation, and release of inflammatory cytokines (ie, tumor necrosis factor [TNF] and reduced systemic endotoxemia).21 One meta-analysis by Heyland et al showed a trend toward reduced infectious morbidity (RR = 0.66; 95% CI 0.36-1.22; P =.08) and mortality (RR = 0.52; 95% CI 0.25-1.08; P = .08),21 while a second by Marik and Zaloga showed significant reductions in infectious morbidity (RR = 0.45; 95% CI 0.30-0.66; P = .00006) and hospital length of stay (mean 2.2 days, 95% CI 0.81-3.63 days; P = .001) with early EN compared to delayed feedings.62 See Table 4 . 63-72 A5. In the setting of hemodynamic compromise (patients requiring significant hemodynamic support including high dose catecholamine agents, alone or in combination with large volume fluid or blood product resuscitation to maintain cellular perfusion), EN should be withheld until the patient is fully resuscitated and/or stable. (Grade: E) Rationale. At the height of critical illness, EN is being provided to patients who are prone to GI dysmotility, sepsis, and hypotension and thus are at increased risk for subclinical ischemia/reperfusion injury involving the intestinal microcirculation. Ischemic bowel is a rare complication of EN, occurring in <1% of cases.73,74 EN-related ischemic bowel has been reported most often in the past with use of surgical jejunostomy tubes. However, more recently, this complication has been described with use of nasojejunal tubes.75 EN intended to be infused into the small bowel should be withheld in patients who are hypotensive (mean arterial blood pressure <60 mm Hg), particularly if clinicians are initiating use of catecholamine agents (eg, norepinephrine, phenylephrine, epinephrine, dopamine) or escalating the dose of such agents to maintain hemodynamic stability. EN may be provided with caution to patients into either the stomach or small bowel on stable low doses of pressor agents,76 but any signs of intolerance (abdominal distention, increasing nasogastric tube output or gastric residual volumes, decreased passage of stool and flatus, hypoactive bowel sounds, increasing metabolic acidosis and/or base deficit) should be closely scrutinized as possible early signs of gut ischemia. A6. In the ICU patient population, neither the presence nor absence of bowel sounds nor evidence of passage of flatus and stool is required for the initiation of enteral feeding. (Grade: B) Rationale. The literature supports the concept that bowel sounds and evidence of bowel function (ie, passing flatus or stool) are not required for initiation of enteral feeding. GI dysfunction in the ICU setting occurs in 30%-70% of patients depending on the diagnosis, premorbid condition, ventilation mode, medications, and metabolic state.77 Proposed mechanisms of ICU and postoperative GI dysfunction can be separated into 3 general categories: mucosal barrier disruption, altered motility and atrophy of the mucosa, and reduced mass of GALT. Bowel sounds are only indicative of contractility and do not necessarily relate to mucosal integrity, barrier function, or absorptive capacity. Success at attaining nutrition goals within the first 72 hours ranges from 30% to 85%. When ICU enteral feeding protocols are followed, rates of GI tolerance in the range of 70%-85% can be achieved.76 Ten randomized clinical trials,63-72 the majority in surgical critically ill patients, have reported feasibility and safety of enteral feeding within the initial 36-48 hours of admission to the ICU. The grade of this recommendation is based on the strength of the literature supporting A3, where patients in the experimental arm of the above mentioned studies were successfully started on EN within the first 36 hours of admission (regardless of clinical signs of stooling, flatus, or borborygmi). See Table 4 . 63-72 A7. Either gastric or small bowel feeding is acceptable in the ICU setting. Critically ill patients should be fed via an enteral access tube placed in the small bowel if at high risk for aspiration or after showing intolerance to gastric feeding. (Grade: C) Withholding of enteral feeding for repeated high gastric residual volumes alone may be sufficient reason to switch to small bowel feeding (the definition for high gastric residual volume is likely to vary from one hospital to the next, as determined by individual institutional protocol). (Grade: E) (See guideline D4 for recommendations on gastric residual volumes, identifying high risk patients, and reducing chances for aspiration.) Rationale. Multiple studies have evaluated gastric vs jejunal feeding in various medical and surgical ICU settings. One level II study comparing gastric vs jejunal feeding showed significantly less gastroesophageal reflux with small bowel feeding.78 In a nonrandomized prospective study using a radioisotope in an enteral formulation, esophageal reflux was reduced significantly with a trend toward reduced aspiration as the level of infusion was moved from the stomach down through the third portion of the duodenum.79 Three meta-analyses have been published comparing gastric with post-pyloric feeding in the ICU setting.80-82 Only 1 of these meta-analyses showed a significant reduction in ventilator-associated pneumonia with post-pyloric feeding (RR = 0.76; 95% CI 0.59-0.99; P = .04),82 an effect heavily influenced by 1 study by Taylor et al.23 With removal of this study from the meta-analysis, the difference was no longer significant. The 2 other meta-analyses (which did not include the Taylor study) showed no difference in pneumonia between gastric and post-pyloric feeding.80,81 While 1 showed no difference in ICU length of stay,80 all 3 meta-analyses showed no significant difference in mortality between gastric and post-pyloric feeding.80-82 See Table 5 . 23,68,78,83-91 B1. If early EN is not feasible or available the first 7 days following admission to the ICU, no nutrition support therapy (ie, STD therapy) should be provided. (Grade: C) In the patient who was previously healthy prior to critical illness with no evidence of protein-calorie malnutrition, use of PN should be reserved and initiated only after the first 7 days of hospitalization (when EN is not available). (Grade: E) Rationale. These 2 recommendations are the most controversial in these guidelines, are influenced primarily by 2 meta-analyses, and should be interpreted very carefully in application to patient care.24,92 Both meta-analyses compared use of PN with STD therapy (where no nutrition support therapy was provided). In critically ill patients in the absence of pre-existing malnutrition (when EN is not available), Braunschweig et al aggregated 7 studies93-99 and showed that use of STD therapy was associated with significantly reduced infectious morbidity (RR = 0.77; 95% CI 0.65-0.91; P <.05) and a trend toward reduced overall complications (RR = 0.87; 95% CI 0.74-1.03; P not provided) compared to use of PN.24 In the same circumstances (critically ill, no EN available, and no evidence of malnutrition), Heyland et al92 aggregated 4 studies96,97,100,101 and showed a significant increase in mortality with use of PN (RR = 0.1.78; 95% CI 1.11-2.85; P < .05) and a trend toward greater rate of complications (RR = 2.40; 95% CI 0.88-6.58; P not provided), when compared to STD therapy. See Table 6 . 93-129 With increased duration of severe illness, priorities between STD therapy and PN become reversed. Sandstrom et al first showed that after the first 14 days of hospitalization had elapsed, continuing to provide no nutrition therapy was associated with significantly greater mortality (21% vs 2%, P < .05) and longer hospital length of stay (36.3 days vs 23.4 days, P < .05), when compared respectively to use of PN.96 The authors of both metaanalyses speculated as to the appropriate length of time before initiating PN in a patient on STD therapy who has not begun to eat spontaneously (Braunschweig recommending 7-10 days, Heyland recommending 14 days).24,92 Conflic ting data were reported in a Chinese study of patients with severe acute pancreatitis. In this study, a significant step-wise improvement was seen in each clinical outcome parameter (hospital length of stay, pancreatic infection, overall complications, and mortality) when comparing patients randomized to STD therapy vs PN vs PN with parenteral glutamine, respectively.121 Because of the discrepancy, we attempted to contact the authors of this latter study to get validation of results but were unsuccessful. The final recommendation was based on the overall negative treatment effect of PN over the first week of hospitalization seen in the 2 metaanalyses.24,92 Although the literature cited recommends withholding PN for 10-14 days, the Guidelines Committee expressed concern that continuing to provide STD therapy (no nutrition support therapy) beyond 7 days would lead to deterioration of nutrition status and an adverse effect on clinical outcome. B2. If there is evidence of protein-calorie malnutrition on admission and EN is not feasible, it is appropriate to initiate PN as soon as possible following admission and adequate resuscitation. (Grade: C) Rationale. In the situation where EN is not available and evidence of protein-calorie malnutrition is present (usually defined by recent weight loss of >10%-15% or actual body weight <90% of ideal body weight), initial priorities are reversed and use of PN has a more favorable outcome than STD therapy. See Table 6 . 93-129 In the Heyland meta-analysis, use of PN in malnourished ICU patients was associated with significantly fewer overall complications (RR = 0.52; 95% CI 0.30-0.91; P < .05) than STD therapy.92 In the Braunschweig meta-analysis, STD therapy in malnourished ICU patients was associated with significantly higher risk for mortality (RR = 3.0; 95% CI 1.09-8.56; P < .05) and a trend toward higher rate of infection (RR = 1.17; 95% CI 0.88-1.56; P not provided) compared to use of PN.24 For these patients, when EN is not available, there should be little delay in initiating PN after admission to the ICU. B3. If a patient is expected to undergo major upper GI surgery and EN is not feasible, PN should be provided under very specific conditions: If the patient is malnourished, PN should be initiated 5-7 days preoperatively and continued into the postoperative period. (Grade: B) PN should not be initiated in the immediate postoperative period but should be delayed for 5-7 days (should EN continue not to be feasible). (Grade: B) PN therapy provided for a duration of <5-7 days would be expected to have no outcome effect and may result in increased risk to the patient. Thus, PN should be initiated only if the duration of therapy is anticipated to be ≥7 days. (Grade: B) Rationale. One population of patients that has shown more consistent benefit of PN over STD involve those patients undergoing major upper GI surgery (esophagectomy, gastrectomy, pancreatectomy, or other major reoperative abdominal procedures), especially if there is evidence of preexisting protein-calorie malnutrition and the PN is provided under specific conditions.24,92 Whereas critically ill patients in the Heyland meta-analysis experienced increased mortality with use of PN compared to STD therapy (see rationale for guideline B1 above), surgical patients saw no treatment effect with PN regarding mortality (RR = 0.91; 95% CI 0.68-1.21; P = NS).92 Critically ill patients experienced a trend toward increased complications, while surgical patients saw significant reductions in complications with use of PN regarding mortality (RR = 2.40; 95% CI 0.88-6.58; P < .05).92 These benefits were noted when PN was provided preoperatively for a minimum of 7-10 days and then continued through the perioperative period. In an earlier meta-analysis by Detsky et al130 comparing perioperative PN with STD therapy, only seven95,98,102,103,107,110,111 out of 14 studies94,100,104,106,108,109,112 provided PN for ≥7 days.130 As a result, only 1 study showed a treatment effect95 and the overall meta-analysis showed no statistically significant benefit from PN.130 In contrast, a later meta-analysis by Klein et al131 aggregated the data from 13 studies,95,98,103,105,111,113-120 all of which provided PN for ≥7 days.131 Six of the studies showed significant beneficial treatment effects from use of PN,95,103,105,111,115,120 with the pooled data from the overall meta-analysis showing a significant 10% decrease in infectious morbidity compared to STD therapy.131 See Table 6 . 93-129 It is imperative to be aware that the beneficial effect of PN is lost if given only postoperatively. Aggregation of data from 9 studies that evaluated routine postoperative PN93,94,96,99-101,104,109,122 showed a significant 10% increase in complications compared to STD therapy.131 Because of the adverse outcome effect from PN initiated in the immediate postoperative period, Klein et al recommended delaying PN for 5-10 days following surgery if EN continues not to be feasible.131 C1. The target goal of EN (defined by energy requirements) should be determined and clearly identified at the time of initiation of nutrition support therapy. (Grade: C) Energy requirements may be calculated by predictive equations or measured by indirect calorimetry. Predictive equations should be used with caution, as they provide a less accurate measure of energy requirements than indirect calorimetry in the individual patient. In the obese patient, the predictive equations are even more problematic without availability of indirect calorimetry. (Grade: E) Rationale. Clinicians should clearly identify the goal of EN, as determined by energy requirements. Over 200 predictive equations (including Harris-Benedict, Scholfield, Ireton-Jones, etc) have been published in the literature.132 Energy requirements may be calculated either through simplistic formulas (25-30 kcal/kg/d), published predictive equations, or the use of indirect calorimetry. Calories provided via infusion of propofol should be considered when calculating the nutrition regimen. While it is often difficult to provide 100% of goal calories by the enteral route, studies in which a protocol was used to increase delivery of EN have shown that delivering a volume of EN where the level of calories and protein provided is closer to goal improves outcome.133,134 This recommendation is supported by two level II studies in which those patients who by protocol randomization received a greater volume of EN experienced significantly fewer complications and less infectious morbidity,23 as well as shorter hospital lengths of stay, and a trend toward lower mortality135 than those patients receiving lower volume. C2. Efforts to provide >50%-65% of goal calories should be made in order to achieve the clinical benefit of EN over the first week of hospitalization. (Grade: C) Rationale. The impact of early EN on patient outcome appears to be a dose-dependent effect. “Trickle” or trophic feeds (usually defined as 10-30 mL/h) may be sufficient to prevent mucosal atrophy but may be insufficient to achieve the usual endpoints desired from EN therapy. Studies suggest that >50%-65% of goal calories may be required to prevent increases in intestinal permeability in burn and bone-marrow transplant patients, to promote faster return of cognitive function in head injury patients, and to improve outcome from immune-modulating enteral formulations in critically ill patients.5,23,133,136 This recommendation is supported by one level II23 and one level III study136 where increases in the percent goal calories infused from a range of 37%-40% up to 59%-64% improved clinical outcome. C3. If unable to meet energy requirements (100% of target goal calories) after 7-10 days by the enteral route alone, consider initiating supplemental PN. (Grade: E) Initiating supplemental PN prior to this 7-10 day period in the patient already receiving EN does not improve outcome and may be detrimental to the patient. (Grade: C) Rationale. Early on, EN is directed toward maintaining gut integrity, reducing oxidative stress, and modulating systemic immunity. In patients already receiving some volume of EN, use of supplemental PN over the first 7-10 days adds cost137,138 and appears to provide no additional benefit.42,137-140 In 1 small study in burn patients, EN supplemented with PN was associated with a significant increase in mortality (63% vs 26%, P < .05) when compared respectively to hypocaloric EN alone.138 See Table 7 . 42,137-140 As discussed in guideline B1, the optimal time to initiate PN in a patient who is already receiving some volume of enteral feeding is not clear. The reports by Braunschweig et al and Sandstrom et al infer that after the first 7-10 days, the need to provide adequate calories and protein is increased in order to prevent the consequences of deterioration of nutrition status.24,96 At this point, if the provision of EN is insufficient to meet requirements, then the addition of supplemental PN should be considered. C4. Ongoing assessment of adequacy of protein provision should be performed. The use of additional modular protein supplements is a common practice, as standard enteral formulations tend to have a high non-protein calorie:nitrogen ratio. In patients with body mass index (BMI) <30, protein requirements should be in the range of 1.2-2.0 g/kg actual body weight per day, and may likely be even higher in burn or multi-trauma patients. (Grade: E) Rationale. In the critical care setting, protein appears to be the most important macronutrient for healing wounds, supporting immune function, and maintaining lean body mass. For most critically ill patients, protein requirements are proportionately higher than energy requirements and therefore are not met by provision of routine enteral formulations. The decision to add protein modules should be based on an ongoing assessment of adequacy of protein provision. Unfortunately in the critical care setting, determination of protein requirements is difficult