Nutrition Considerations in the Intensive Care Unit

Enteral nutrition dates back to 3500 bc, when the ancient Egyptians would administer enemas of wine, milk, whey, wheat, and barley.1 Nearly 5500 years later, Dr Stanley Dudrick successfully nourished beagle puppies with total parenteral nutrition (TPN).2 The practice of nutrition has evolved greatly over the years, yet many unanswered questions regarding nutrition in critically ill patients remain. Malnutrition affects up to 78% of intensive care unit (ICU) patients, leading to increased morbidity, mortality, and hospital-related costs.3 This review explores the considerations and controversies of feeding adult patients in the ICU.In critically ill patients, caloric intake is either limited or contraindicated due to the high rates of intubation, lack of appetite, and nothing-by-mouth status. This limitation in intake creates a dilemma because many disease states in ICU patients are extremely catabolic. Up to 1 kg of muscle mass per day is thought to be lost within the first 10 days of an ICU stay, regardless of nutrition administration.4 Enteral and parenteral nutrition have become ubiquitous in health care because they help overcome these acute challenges.Multiple types of feeding tubes, such as nasogastric, orogastric, nasojejunal, percutaneous endoscopic, and surgically placed tubes, exist to provide enteral nutrition. Many types of enteral nutrition formulas, with varying recommendations for use, are available.5,6 The goal of enteral nutrition is to bypass common ICU barriers to nutrition initiation, such as mechanical ventilation or a lack of appetite. Enteral nutrition is commonly preferred if the patient’s intestinal tract is functional.5,6Parenteral nutrition allows for nutrition administration directly into the vasculature through venous access. Total parenteral nutrition bypasses the gastrointestinal tract, which is preferable when bowel rest is needed. Common indications for TPN include small-bowel obstruction, chronic severe diarrhea and/or vomiting, and recent bowel surgery with complications.7 Limitations of parenteral nutrition are numerous; it is associated with increases in infection rate, ICU length of stay, and costs.5,8Continued catabolism without appropriate caloric intake can lead to worse outcomes, including prolonged hospital stays, fewer ventilator-free days, and increased infection rates.9 The more critical the disease state, the more likely are barriers to initiating nutrition. Patients with more severe disease states may benefit from early nutrition, particularly enteral nutrition, making it essential to overcome barriers to enteral nutrition initiation.Many hypotheses exist as to why initiation of enteral feeding is critical for patients in the ICU. Nutrition plays a significant role in the health and viability of the gut. In the absence of enteral nutrition, alterations to the intestinal lining occur.9 In disease states such as sepsis, tight junctions in the epithelial lining are impaired, which in the absence of luminal nutrition can lead to overgrowth of virulent pathogens and increased oxidative stress.9 Lack of luminal nutrition also decreases mesenteric blood flow, which promotes ischemia and reperfusion injury, synergistically impacting inflammation. Introduction of nutrition can lead to gut-associated lymphoid tissue stimulation, early gut perfusion, and a dampened inflammatory response.9More than 3 decades of research support early enteral nutrition initiation. The American Society for Enteral and Parenteral Nutrition (ASPEN), in their 2016 guidelines for nutrition support in adult critically ill patients, recommend initiating enteral nutrition within 24 to 48 hours of ICU admission if possible.5 Some studies have advocated for earlier initiation. Yu et al10 conducted a randomized, interventional clinical trial comparing enteral nutrition initiation via nasojejunal tube at admission with nutrition initiation via nasogastric tube at 24 to 48 hours after admission. They found that serum albumin and prealbumin levels were significantly higher and that ICU stays and mechanical ventilation duration were significantly shorter in the group who received earlier nutrition initiation. The researchers excluded patients who had hemodynamic instability or contraindications to enteral nutrition.Of possibly greater significance is the risk of early overfeeding in ICU patients. Critical illness leads to a catabolic state. Although the absence of nutrition can precipitate worse outcomes, such as prolonged ICU stays, overfeeding may be detrimental. In a retrospective analysis of 1171 patients receiving mechanical ventilation, Zusman et al11 described a U-shaped relationship between 60-day mortality and caloric intake, associating both high and low caloric intake with increased mortality. This finding was further supported by results of a meta-analysis by Choi et al,12 who showed in a subgroup analysis that targeting 33% to 70% of caloric goals in the first 5 to 7 days of ICU admission resulted in lower mortality than did targeting 0% to 33% or 71% to 100% of caloric goals.Critical care clinicians have the dilemma of balancing the initiation of enteral nutrition with the risks of overfeeding. Clinical dietitians are of utmost importance because their input can help clinicians navigate not only the titration but also the different mechanisms for caloric optimization for patients in the ICU. Table 1provides further recommendations on the initiation of enteral nutrition.The initiation of enteral nutrition is optimal in most patients, but certain disease states pose challenges. An enterocutaneous fistula (ECF) is a complication that can result from surgery, malignancy, inflammatory bowel disease, radiation therapy, infection, or diverticular disease.15 An ECF is an abnormal connection between the enteral tract and skin and is associated with significant morbidity and mortality.15,16 Appropriate management is paramount to clinical success, and nutrition optimization is critical.Enteral feeding is highly recommended in patients with an ECF because early enteral nutrition initiation is associated with earlier fistula closure and lower recurrence rates15 in addition to its general significant benefit for patients with critical illness. An ECF can be classified according to its 24-hour output. A high-output fistula expels greater than 500 mL of fluid per day.16 Enteral feeding often increases the output of this type of ECF, leading to abnormalities in electrolyte levels and an inability to achieve caloric goals.13 The inability to manage electrolyte abnormalities and volume loss may prohibit the use of enteral nutrition in these patients.Total parenteral nutrition is commonly used in patients with high-output fistulas.14 Total parenteral nutrition provides nutritional support without worsening ECF output, potentially increasing the chance of spontaneous closure. The 2016 ASPEN–Federación Latino Americana de Terapia Nutricional, Nutrición Clínica y Metabolismo (FELANPE) guidelines for nutrition support of adult patients with ECF recommend using TPN for patients with a high-output ECF.13 Protein intake should be prioritized because of the patient’s extreme inflammatory state and protein loss through the ECF. The protein administration target is as high as 2.5 g/kg/d.13 Enteral nutrition may be considered while TPN is ongoing, although careful monitoring of ECF output volume is required.Another feeding modality is fistuloclysis, or providing nutrition through the fistula itself.15 Patients with a distal ECF have a short effective bowel length for nutrition absorption, and fistuloclysis may be ineffective. The ECF is also more likely to become epithelialized with the use of fistuloclysis, limiting the chance of spontaneous closure.14 The 2016 ASPEN-FELANPE guidelines therefore suggest using fistuloclysis in patients in whom spontaneous closure is unlikely and nutrition absorption distal to the ECF is expected.13Important to the management of an ECF is surgical closure, which is beyond the scope of this article. The surgical procedure is often delayed for months16 because an earlier surgical intervention may create further injury and harm while preventing possible spontaneous closure. Optimizing nutritional status during this time improves surgical outcomes for patients with an ECF, making nutrition an important aspect of a patient’s care.16 Table 1provides further recommendations for nutrition management in patients with a fistula.Enteral nutrition is often given through a feeding tube in the ICU, and concomitant medication administration through the tube is common. Medications are formulated for administration to a patient with normal gastrointestinal anatomy. Once a feeding tube is placed, the exit site of the tube must be considered. The pharmacokinetics and adverse effect profile of medications may be altered depending on the type of feeding tube placed. Nasojejunal tubes, for example, bypass multiple levels of the gastrointestinal tract.Each medication must be assessed individually. Medications such as aspirin that are absorbed at the level of the stomach may have decreased bioavailability if administered distal to the stomach.17 The duodenum is also essential for the absorption of certain medications. Iron is primarily absorbed at the duodenum after gastric dissolution.18 If iron is administered through a nasojejunal or percutaneous endoscopic jejunostomy tube, bioavailability is limited. Tertiary resources that curate information on absorption and site of administration for different medications, such as the Handbook of Drug Administration via Enteral Feeding Tubes,18 are available. These resources, institution-specific protocols, and the pharmacist on the care team are helpful in mitigating diminished medication efficacy and adverse effects.Another consideration for feeding tube medication administration is whether a medication can be crushed or diluted. Extended-release and film-coated formulations are examples of medications that should not be crushed and administered through a feeding tube.17 Crushing extended-release medications may result in an increased rate of medication absorption, leading to supratherapeutic levels and potential patient harm or even death.17 If a medication can be crushed or is a liquid, the osmolality and viscosity of the resulting liquid may be too high, which can lead to clogging of the tube or patient discomfort. Care should be taken in referencing the package insert or contacting the pharmacist for further information on appropriate crushing and dilution directions.17 Table 1provides further recommendations on medication administration in the ICU.Hemodynamic instability is ubiquitous in patients in the ICU and is a common reason for ICU admission. A patient may have hemodynamic instability for a multitude of reasons, and guidelines address most of these pathologies. In severe instances, vasoactive support from vasopressors and inotropes are required to maintain perfusion of tissues. Whether enteral nutrition should be started in patients requiring vasopressors is unclear. Guidelines currently recommend initiating enteral nutrition after the achievement of hemodynamic stability.5,6 A main reason for the delay of enteral nutrition initiation is the concern for mesenteric ischemia and nonocclusive bowel necrosis with nutrition initiation. This concern could preclude starting enteral nutrition within the recommended 48-hour window.19No randomized controlled trials have directly compared outcomes in patients who are receiving enteral nutrition and vasopressors versus those who are not receiving both enteral nutrition and vassopressors. Correlating a causal relationship between bowel ischemia and feeding in patients receiving pressors is difficult. However, observational data suggest that feeding patients receiving pressors is safe and improves outcomes. Khalid et al20 prospectively collected data from 1174 patients receiving mechanical ventilation and vasopressor support. Patients were stratified as those who received or did not receive enteral nutrition within 48 hours of intubation. Early enteral nutrition was associated with improved mortality even after propensity score matching. Unfortunately, rates of bowel ischemia and vasopressor doses were not reported.Data from randomized trials suggest that enteral feeding is safe for patients receiving vasopressors. Neither the EDEN (Early Versus Delayed Enteral Feeding to Treat People With Acute Lung Injury or Acute Respiratory Distress Syndrome) trial nor the PermiT (Permissive Underfeeding Versus Target Enteral Feeding in Adult Critically Ill Patients) trial exclude patients receiving vasopressors, but the investigators did set exclusion criteria for patients receiving higher doses.21,22 Pressors were given to 38% of patients in the EDEN trial and 55% of patients in the PermiT trial. Patients were excluded if they received norepinephrine doses greater than 30 μg/min and 0.4 μg/kg/min, respectively, in the EDEN and PermiT trials. Although these trials examined permissive underfeeding versus trophic feeding, the PermiT trial investigators completed a subgroup analysis and found no difference in mortality according to receipt of vasopressor therapy. The results of both trials suggested that outcomes were similar for patients receiving enteral nutrition regardless of pressor administration. Unfortunately, the EDEN trial investigators did not complete a subgroup analysis for patients receiving vasopressor therapy.Results of the NUTRIREA-2 (Impact of Early Enteral vs Parenteral Nutrition on Mortality in Patients Requiring Mechanical Ventilation and Catecholamines) trial, a randomized, controlled, multicenter, open-label, parallel-group study, were published in Lancet in 2018.23 In the NUTRIREA-2 trial, completed at 44 French ICUs, 2410 patients receiving mechanical ventilation and vasopressor support were randomized to receive enteral nutrition exclusively during the first 7 days after admission or to receive parenteral nutrition exclusively during the first 3 days after admission. The results of the NUTRIREA-2 trial suggested that enteral nutrition was not safe in patients receiving high-dose vasopressors. The authors reported that the mean dose of norepinephrine used during the study was 0.53 μg/kg/min. Although rates of bowel ischemia were low, the rate in the enteral nutrition group was 4 times higher than in the parenteral nutrition group, a significant difference. The NUTRIREA-2 trial was the first randomized trial to show that administering enteral nutrition targeted to caloric goals was harmful to patients requiring high-dose vasopressors. Whether a more conservative trophic feeding approach could have been beneficial in these patients is not known.Some guidance may be drawn from retrospective data. Mancl and Muzevich24 found an association between enteral nutrition tolerability and vasopressor dose, suggesting that norepinephrine doses of less than 12.5 μg/min (0.157 μg/kg/min) were more likely to be associated with nutrition tolerance compared with doses above 12.5 μg/min (0.157 μg/kg/min). Merchan et al25 found similar results in patients receiving mechanical ventilation, associating early enteral nutrition tolerability with norepinephrine-equivalent doses of less than 0.14 μg/kg/min.Future randomized controlled trials evaluating doses of norepinephrine that are safe for enteral feeding and whether certain vasopressor agents should be avoided would be beneficial. Nutrition administration becomes more difficult in patients who require more than 1 vasopressor agent. Table 2includes recommendations for feeding patients receiving vasopressor support.Nutrition in critically ill patients is important yet often overlooked. Early enteral nutrition should be prioritized in every ICU patient. Many modalities and formulations are used for enteral nutrition, and each has a specific use. When nutrition is initiated, it should be administered slowly to avoid overfeeding during the first week of a patient’s ICU stay. Enteral nutrition is preferred over parenteral nutrition, but TPN may be started earlier in patients who are hemodynamically unstable or extremely malnourished. If enteral nutrition is started in a patient receiving vasopressors, dose escalation should be done with care and nutrition delivery should be paused as necessary for patients receiving norepinephrine or equivalent doses of greater than 0.15 μg/kg/min.