A middle aged man is admitted with abdominal pain and vomiting. He has a history of alcohol excess. A CT scan shows evidence of pancreatic necrosis. Supportive care is initiated and an NGT placed for supplementary enteral nutrition. After 3 days, he is referred to ICU as his oxygen requirements have increased and he is requiring non-invasive ventilation. It is noted that he has had very large gastric aspirates. Parenteral nutrition is commenced at this point.

What is the evidence for enteral versus parenteral feed as a source of nutrition in critically ill patients?

Dave Slessor

Malnutrition is a state in which a deficiency of energy, protein and/or other nutrients causes measurable adverse effects on tissue/body form, composition, functional or clinical outcome (1). In critically ill patients the inflammatory response, metabolic stress and bed rest all lead to catabolism and malnutrition (2). Malnutrition causes impaired immune responses and wound healing, as well as reduced muscle strength and electrolyte disturbances (1). Malnutrition has been demonstrated to lead to increased ventilator requirements as well as increased morbidity and mortality for patients in the ICU (3).

Therefore it is recommended that patients in the ICU, who are unable to take a normal oral diet, should still be fed (4). The feed should include fat, carbohydrates, protein, electrolytes, minerals, micronutrients and fluid (1). EN is preferred to PN (4) as in comparison it is associated with a significant decrease in infectious complications (relative risk=0.64,95% C.I 0.47-0.87, P=0.0004) but with no difference in mortality, length of ventilation or hospital stay (3).

It is difficult to define the nutritional requirements of critically ill patients as there is a large variation in energy expenditure secondary to the underlying illness as well as treatment given (2). Equations have been developed to calculate basal metabolic rate, to which corrections are added to cover expenditure secondary to the metabolic stress and variations in activity level (1). However, for most patients 20-30kcal/kg/day is likely to be adequate (1).

Energy expenditure can be measured on an individual process using indirect calorimetry (IC) (2). In a non-blinded randomised controlled trial (RCT) Singer reported a trend to improved mortality in patients that had their energy goal set by IC compared to patients whose energy requirements were set at a standard 25kcal/kg/day (32.3% vs. 47.7%, P=0.058) (5). However, in the IC group the % of target feed that was given was 106% compared to only 75% in the standard group. The large difference in target achieved was greater than the day-to-day variation in the target set as recommended by IC. This bias invalidates the results.

Whether we should be targeting feeds based on matching energy expenditure has not been determined. Overfeeding has been associated with worse glycaemic control, liver function, infections, and outcome,(2). Whereas Villet demonstrated that patients that were underfed and had a higher energy deficit had a longer ICU stay, and more infectious complications (6). However, this was an observational study and therefore the results may be explained by confounders such as severity of illness. Where as in a RCT comparing feeding at 60-70% with 90-100% of calculated requirements, patients in the hypocaloric group had a lower hospital mortality (30% vs. 42.65%, Relative risk 0.71 95%C.I 0.5-0.99, P=0.04) (7). The actual target feed that was achieved in the hypocaloric and control group was 59% and 71% respectively. This was a single centre study. Therefore before accepting that the small difference in feed given led to a mortality benefit we would require further studies to confirm these findings.

If patients are unable to tolerate EN then PN should be started (4). The timing of when to start PN is controversial. To try and answer this question Casaer compared the addition of PN at day 3 vs. day 8 to achieve full target calorie intake in a RCT of over 4000 patients (8). Patients in the early group, when compared to the late group, had a significantly higher rate of new infection (26.2% vs. 2.8%, P=0.008), a longer ICU length of stay (LOS) (median 4 vs. 3 days P=0.02), and a higher rate of patients requiring mechanical ventilation (MV) >2 days (40.2% vs. 36.3%, P=0.006). There was no difference in mortality. In a post-hoc sub-group analysis of patients who had contraindications to early EN, the rate of infection was lower in the late PN group (29.2% vs. 40.2%, P=0.01) and the number of patients discharged alive within eight days was higher (OR 1.749, 95%C.I. 1.141-2.683), when compared to the early PN group.

Heidegger found conflicting results when he compared the addition of PN to EN from day 3-8 in patients who were receiving less than 60% of their target feed at day three compared to continuing EN alone (9). The rate of nosocomial infection from day 8-28 was significantly lower in the PN+EN group compared to the EN alone group (27% vs. 38%, hazard ratio 0.65, 95%C.I. 0.43-0.97). There was no difference in mortality or LOS although it was not powered for this.

The reasons for differences found between the studies of Casaer and Heidegger may be due to differences in populations studied. In the Casaer study 50% of patients stayed less than three days and 61% underwent cardiac surgery. These patients may have had minimal nutritional deficiencies. Where as in the Heidegger study all patients were expected to stay for at least five days and only 13% had cardiac surgery. In the Casaer trial patients in the early PN were given a high glucose load for the first two days which may have contributed to development of infections.

More recently (10) the early PN trial compared standard care vs. providing PN within 24 hours of admission to patients with short term relative contraindications to EN. It was a multi-centre RCT with 1372 randomised patients. In the standard care group patients remained unfed for a mean of 2.8 days before starting EN, PN or both. In the early PN group, PN was started at mean of 44 minutes post randomisation. There was no difference in 60 day mortality or infection rates. Patients in the early PN group had significantly fewer days of MV compared to the standard group (-0.47 days per 10 patient* ICU days, 95% C.I. -0.82 to -0.11, P=0.01). Whether this is clinically significant is questionable especially as there was no statistical difference in ICU or hospital stay.

---

References

1. National Collaborating Centre for Acute Care. Nutrition support in adults Oral nutrition support, enteral tube feeding and parenteral nutrition. National Collaborating Centre for Acute Care, London [Internet]. 2006.
2. Berger MM, Pichard C. Best timing for energy provision during critical illness. Critical Care. 2012;16(2):215.
3. Rupinder Dhaliwal R, Heyland DK. Does enteral nutrition compared to parenteral nutrition result in better outcomes in critically ill adult patients? A systematic review of the literature. Nutrition. 2004;20:843-8.
4. Canadian Clinical Practice Guidelines Committee, Critical Care Nutrition. 
5. Singer P, Anbar R, Cohen J, Shalita-Chesner M, Lev S, et al. The tight calorie control study (TICACOS): a prospective, randomized, controlled pilot study of nutritional support in critically ill patients. Intensive care medicine. 2011;37(4):601-9.
6. Villet S, Chiolero RL, Revelly J-P, Cayeux RN M-C, Delarue J, et al. Negative impact of hypocaloric feeding and energy balance on clinical outcome in ICU patients. Clinical Nutrition. 2005;24(4):502-9.
7. Arabi YM, Tamim HM, Al-Dawood A, Al-Sultan M, Sakkijha MH, et al. Permissive underfeeding and intensive insulin therapy in critically ill patients: a randomized controlled trial. The American journal of clinical nutrition. 2011;93(3):569-77.
8. Casaer MP, Mesotten D, Hermans G, , Schetz M, Meyfroidt G, et al. Early versus Late Parenteral Nutrition in Critically Ill Adults. New England Journal of Medicine. 2011;365(6):506-17.
9. Heidegger CP, Berger MM, Zingg W, Darmon P, Costanza MC, et al. Optimisation of energy provision with supplemental parenteral nutrition in critically ill patients: a randomised controlled clinical trial. The Lancet. 2013.
10. Doig Gs SFSEA, et al. Early parenteral nutrition in critically ill patients with short-term relative contraindications to early enteral nutrition: A randomized controlled trial. JAMA. 2013:1-9.

---