Special Feature

Nutritional Support for the Acutely Critically Ill Patient

By Saadia R. Akhtar MD, MSc, Idaho Pulmonary Associates, Boise, is Associate Editor for Critical Care Alert

Dr. Akhtar reports no financial relationship to this field of study.

Although it is understood that good nutrition is essential for normal immune function, wound healing and maintenance of muscle mass and function, and is likely beneficial for overall recovery from the highly catabolic state of acute critical illness, it is less clear when and how best to deliver adequate nutritional support in practice. Availability of robust data is limited for a variety of reasons. It would be unethical to compare nutrition to no nutrition, and thus such a study has never been done. Furthermore, intensive care unit patients are generally quite heterogeneous in terms of, for instance, admitting diagnoses, severity of illness, co-morbidities and baseline nutritional status. This makes it challenging to design studies and enroll enough subjects within subgroups to achieve the power necessary to evaluate important clinical outcomes. It may also make comparison between studies unfeasible.

Nevertheless, several scientific organizations have developed (often differing) practice guidelines and position papers on nutrition in the ICU: the American College of Chest Physicians, American Society for Parenteral and Enteral Nutrition, Canadian Critical Care Clinical Practice Guidelines Committee, and European Society for Clinical Nutrition and Metabolism, to name just a few.1 Using such statements and recent literature, this brief report will provide an overview of the issues relating to nutritional support for the acutely critically ill patient.

How Much Nutrition Should Be Provided?

The total energy expenditure and caloric requirements of critically ill patients vary greatly, based on their specific illness (for example, caloric needs may be much higher for a patient with septic shock compared to one with a head injury) as well as time spent in the hospital. It is clear that there are potential hazards of overfeeding, such as azotemia, hypercarbia, metabolic acidosis, and hyperglycemia. Similarly, there are risks of severe underfeeding, such as increased infections and mortality. Some authors favor mild to moderate hypocaloric nutrition to avoid the risks of overfeeding; however, there are no data to support or negate this recommendation.2

An estimated total energy expenditure of about 25 ± 5 kcal/kg/day is a widely accepted starting point.3 It is suggested that this be provided with about 1.5g/kg/day protein, ≤ 6g/kg/day carbohydrates, and ≤ 2.5kg/day lipids.1 Indirect calorimetry may be used to guide nutritional support, particularly if there are clinical signs concerning for over- or under-feeding. However, there is no specific evidence for or against its utility.

Only 1 study has evaluated the starting volume of enteral feedings.4 Eighty-two patients with head injury were randomized to receive feedings at their goal rate immediately or advance gradually. The former was associated with improved neurological outcomes and decreased infection risk.

What is the Best Route of Administration?

While some debate remains over the benefits, risks, and respective roles of enteral vs parenteral nutrition for acutely critically ill patients, most studies, experts and current practice guidelines recommend the enteral route as the first choice. Enteral nutrition maintains gastrointestinal integrity,5 while parenteral nutrition is clearly associated with gut mucosal atrophy, bacterial translocation, and increased risk of infection. (It is important to note that the risk of infection with parenteral nutrition cannot be fully explained by hyperglycemia.6)

Especially if poorly tolerated, with high gastric residuals, enteral nutrition may increase the risk of hospital- and ventilator-associated pneumonia, and also prolong mechanical ventilation and ICU stays.7 However, these risks can be reduced by elevation of the head of the bed.8 Although advocated by some, the utility of small bowel feeding for decreasing risk of nosocomial pneumonia when compared to gastric feeding remains unclear.9,10

One meta-analysis evaluated 13 randomized studies comparing enteral to parenteral nutrition in mechanically ventilated critically ill patients. Relative risk for any infection was 0.61 for patients receiving enteral nutrition.11 These results were replicated in a second systematic review.12

Overfeeding and hyperglycemia are also potential concerns, particularly with parenteral nutrition. Hyperglycemia is less a concern today with the use of intensive insulin drip protocols.13 It is of interest that enteral nutrition often unintentionally provides hypocaloric feeding, at least initially: several observational studies and surveys have shown that generally only about 50-60% of the target is achieved in the first few days of enteral feeding.14 Withholding lipids with parenteral nutrition or providing hypocaloric parenteral nutrition may decrease infection risk and improve other outcomes as well.15

Finally, according to available evidence, there is no difference in outcomes between using enteral nutrition alone and combining enteral and parenteral nutrition.11

When Should Nutrition be Started?

In the past, enteral feeding was often not initiated until day 5-7 of critical illness. The arguments supporting this included the common occurrence of gastroparesis and decreased bowel motility in acute illness and the presence of "adequate" stores in previously healthy patients. Over time it has become clear that not only does small bowel function remain intact, but lack of feeding may lead to villous atrophy, loss of gastrointestinal integrity and increased bacterial translocation. Multiple animal studies confirm that early enteral feeding reduces markers of ischemic/oxidative injury in various organs, improves gastrointestinal blood flow, decreases bacterial translocation from the gut and even improves wound healing.

The benefits of early enteral feeding have been demonstrated in human clinical trials as well. A 2001 meta-analysis of 15 randomized controlled trials (with 753 patients) compared early (within 36 hours of admission or surgery) to late enteral nutrition in acutely ill patients (post-surgical, head injured, burn and medical ICU).16 Although there was considerable heterogeneity between the studies, infection risk and length of stay were significantly reduced in the early enteral nutrition group (19% vs 41% and mean reduction of 2 days, respectively). There were non-significant trends towards lower risk of non-infectious complications and death as well. Heyland et al's 2003 meta-analysis of 8 trials (including 4 studies not considered by the 2001 work) comparing early (within 48 hours of admission to an intensive care unit) to late enteral feeding arrived at similar results.11

Does the Specific Composition of the Nutritional Support Matter?

Enteral feeding formulas with a variety of additives with potential for immune modulation are available. These additives include arginine, glutamine, omega-3 fatty acids, antioxidants, fiber, and others. There have been several studies of such immunonutrition, without consistent evidence of benefit.3, 11 The largest randomized controlled trial of a formula containing several of these additives evaluated 597 mixed surgical and medical intensive care unit patients and found no differences in infections, lengths of stay or mortality.17

One randomized controlled study compared Oxepa® (enteral formula containing fish oil and antioxidants) to a standard formula in patients with ARDS. Those patients receiving Oxepa® had reduced ventilator time, intensive care unit length of stay and organ failures: there was a non-significant trend towards reduced mortality.18 This feeding formula has also been studied in patients with severe sepsis and septic shock with similar results.19 Both of these studies have been criticized for using a potentially pro-inflammatory, high fat control formula. Furthermore, because Oxepa® contains a variety of substances that may impact the inflammatory response, it is difficult to know which component led to benefit.

Glutamine has been shown to reduce infections, improve wound healing and possibly improve mortality in small studies of trauma and burn patients. This has been demonstrated with glutamine supplementation to parenteral nutrition as well. Glutamine supplementation for larger, heterogeneous intensive care unit populations, however, imparts no benefit.11

Pre-, pro- and synbiotics added to enteral formulations for nutritional support of critically ill patients do not appear to impact clinical outcomes.20

Finally, no outcome differences have been demonstrated for alimental vs whole-protein feeding formulas in critically ill patients. (The former may be associated with increased diarrhea and are certainly appropriate for patients with chronic malabsorption/GI disease such as short bowel syndrome, chronic pancreatitis or other.)11


Limited data suggest that nutritional support should be initiated for all critically ill patients within 24-48 hours of admission. The enteral route is preferred and should be maximally pursued (with use of pro-motility agents or small bowel feeding tubes if needed). Parenteral nutrition should be reserved for those patients unable to receive/tolerate enteral feeds (for instance, bowel obstruction, bowel ischemia, severe ileus). Intensive glucose control should be implemented and patients carefully monitored to avoid over- or severe under-feeding. Consideration could be given to adding glutamine for trauma and burn patients and using Oxepa® for patients with ARDS or septic shock. Finally, protocols and algorithms may facilitate initiation of nutrition in critical care settings and improve clinical outcomes.21


  1. Bozzetti F, Allaria B. Nutritional support in ICU patients: Position of scientific societies. Nestle Nutr Workshop Ser Clin Perform Programme 2003:8:279-292.
  2. Boitano, M. Nutr Clin Pract. 2006:21:617-622.
  3. Bongers T, Griffiths RD. Curr Opin Crit Care. 2006:12:131-135.
  4. Taylor SJ, et al. Crit Care Med. 1999: 27:2525-2531.
  5. Kompan L, et al. Intensive Care Med. 1999:25:157-161.
  6. Heyland DK, et al. JAMA. 1998:280: 2013-2019.
  7. Mentec H, et al. Crit Care Med. 2001: 29:1955-1961.
  8. Drakulovic MB, et al. Lancet. 1999:354: 1851-1858.
  9. Montejo JC, et al. Crit Care Med. 2002:30:796-800.
  10. Marik PE, Zaloga GP. Crit Care. 2003:7:46-51.
  11. Heyland DK, et al, and the Canadian Critical Care Clinical Practice Guidelines Committee. JPEN J Parenter Enteral Nutr. 2003:27:355-373.
  12. Gramlich L, et al. Nutrition. 2004:20:843-848.
  13. van den Berghe G, et al. N Engl J Med. 2001:345:1359-1367.
  14. Heyland DK. JPEN J Parenter Enteral Nutr. 2003:27:74-83.
  15. Battistella FD, et al. J Trauma. 1997:43:52-58.
  16. Marik PE, Zaloga GP. Crit Care Med. 2001:29:2264-2270.
  17. Kieft H, et al. Intensive Care Med. 2005:31:524-532.
  18. Gadek JE, et al. Crit Care Med. 1999:27:1409-1420.
  19. Pontes-Arruda A, et al. Crit Care Med. 2006:34: 2325-2333.
  20. Watkinson PJ, et al. Clinical Nutr. 2007:26:182-192.
  21. Martin CM, et al. Southwestern Ontario Critical Care Research Network. Multicenter, cluster-randomized clinical trial of algorithms for critical-care enteral and parenteral therapy (ACCEPT). CMAJ. 2004:170:197-204.