Skip to main content

Optimal guidance for early nutrition therapy in critical illness?

Most recent RCTs aimed at improving outcome by optimizing nutritional intake early in critical illness by enteral (EN) and/or parenteral (PN) intake did not generate results in support of this approach [1]: feeding at lower doses was as good as enhanced feeding and in some patients populations [1,2,3] even better. In patients suffering from acute lung Injury [2] or at risk of refeeding syndrome [3], providing less artificial nutrition may even have saved lives.

Some nutrition experts attributed these results to “up-to-target-feeding”, defining the energy targets based on calculations rather than measured resting energy expenditure (REE) via indirect calorimetry (IC), and questioned whether the protein intake was sufficient. Many experts claimed that nutrition support was not individualized in most recent RCTs. Also, the lack of functional outcomes may be an issue, as some believe that nutrition may be more likely to preserve function than save lives early in critical disease.

Matilde Allingstrup and co-investigators have now addressed these hypotheses in a RCT of solid methodological design (the EAT trial) [4]. Compared to standard care, daily IC-guided nutrition, with dynamic compensation for protein losses until extubation or ICU discharge, failed to demonstrate a significant improvement in any patient-centered outcomes. The primary functional outcome endpoint—the physical component summary domain of SF-36 at 6 months—was unaffected. Similarly, acute survival and morbidity did not differ between the groups.

The EAT trial is unique in the individualized approach to measuring REE on a daily basis and in compensating for nitrogen losses by modifications based on the nitrogen balance of the previous day in the early goal-directed nutrition (EGDN) arm. A secondary safety mechanism consisted in reducing energy load upon persisting hyperglycemia (defined as insulin needs exceeding 5 IU/h) and nitrogen load when plasma urea rose above 20 mmol/L. This labor-intensive protocol has been executed in a meticulous manner by the very motivated and experienced nutritionists/investigators. Moreover, the study was reported according to the highest RCT standards. The unambiguous primary endpoint was well defined and registered in a public registry without any a posteriori modifications. All outcomes were assessed by blinded experts. The statistical analysis plan took into account the complexities of competing events such as death (see on-line supplement).

Was it this approach, maximally reducing the risk of biased outcome assessment, that resulted in the absence of difference between both strategies? Or was it the overoptimistic power calculation and unexpectedly low mortality as suggested by the authors? Aggregating the outcomes from the three IC-guided nutrition-RCTs [4,5,6] (acknowledging that they differ by design) still does not allow to confirm or exclude a survival benefit or harm (Fig. 1). Or is it the concept of early energy and nitrogen compensation itself that needs to be revisited? Indeed, the two earlier trials of artificial nutrition guided by REE, while providing “sufficient” protein, failed to demonstrate any improvement in patient-centered outcomes either. On the contrary, in both the TICACOS and EAT studies, IC provoked a significantly longer ICU dependency; for the latter, however, this was the result of a post hoc analysis. In the TICACOS trial, this intervention provoked excess infections [5]. In the SPN trial, it reduced the incidence of new infections occurring in the ICU only after censoring the infections provoked early during the nutritional intervention [6].

Fig. 1
figure 1

Absolute risk difference for mortality with metabolic cart guided feeding versus calculated target guided feeding

The question whether IC-guided feeding improves outcome is not only of scientific importance as implementation of IC into an ICU is costly and labor-intensive. One could conclude that REE-guided EGDN was safer than early supplemental parenteral nutrition in the EPaNIC trial. In the latter, “Early-PN” was gradually initiated: initially, 24–48 h of parenteral glucose, completed with amino-acids and lipids thereafter and was targeted at covering estimated needs together with the achieved enteral nutrition, which was started on day 2 in both arms. This “Early-PN” provoked more infections than “Late PN”, starting PN only on day 7 [7]. Yet both studies differed not only by intervention but also by the control arms. “Late-PN” patients—as they received less than 5–10 kcal/kg day—may have benefited from early nutrient restriction, while the EAT control arm still received more than 10 kcal/kg at the first study day. So, in fact, a third “trophic” or “restricted” arm in the EAT trial may have been very informative as a comparator and should be considered in the design of future trials.

An explanation for the lack of beneficial impact on physical function may be in the almost doubling of urinary nitrogen excretion in the EGDN group despite the built-in safety mechanism. This signal suggesting nitrogen breakdown and spill-over has been reported before. In the Nephroprotective trial, additional intravenous amino acids, despite increasing urinary output and glomerular filtration rate, resulted in an important increase in serum urea [8]. Likewise, Jan Gunst revealed that the additional amino acids in EPANIC patients had a modest effect on nitrogen balance and were largely broken down into urinary urea [9]. The metabolic resistance of muscle against early anabolic interventions was illustrated in the same RCT, revealing similar loss of macroscopic and microscopic muscle volume in both study arms despite the important difference in early energy, protein and insulin doses administered [10]. In the Nephroprotective trial, long-term functional outcome was unaffected and the EPaNIC intervention induced more ICU-acquired weakness [10]. Another striking metabolic finding is the higher measured REE in the EGDN arm. Would this reflect thermogenesis and, together with the high insulin needs, a potential metabolic burden of early REE-guided nutrition? In the few RCTs where early enhanced nutrient administration provoked harm, it also provoked an important increase in insulin requirements [2, 3, 7].

Perhaps timing and dosing of energy and protein administration in critical illness should not be guided by the amount of energy burned and protein broken down by the patient. The intensity of spontaneous muscle movements and physical exercise, resolution of early inflammation or insulin resistance might provide more appropriate guidance. These hypotheses should be tested in RCTs powered for evaluation of functional outcomes and conducted at similar methodological standards as the EAT trial. Every such trial, whether it yields neutral, positive or negative results, will contribute to a better understanding of early metabolism in critical illness and ultimately better patient management.


  1. Casaer MP, Van den Berghe G (2014) Nutrition in the acute phase of critical illness. N Engl J Med 370(25):2450–2451

    CAS  PubMed  Google Scholar 

  2. Braunschweig CA, Sheean PM, Peterson SJ, Gomez PS, Freels S, Lateef O et al (2014) Intensive nutrition in acute lung injury: a clinical trial (INTACT). J Parenter Enteral Nutr 39:13–20

    Article  Google Scholar 

  3. Doig GS, Simpson F, Heighes PT, Bellomo R, Chesher D, Caterson ID et al (2015) Restricted versus continued standard caloric intake during the management of refeeding syndrome in critically ill adults: a randomised, parallel-group, multicentre, single-blind controlled trial. Lancet Respir Med 3(12):943–952

    Article  PubMed  Google Scholar 

  4. Allingstrup MJ, Kondrup J, Wiis J, Claudius C, Pedersen U, Rasmussen R et al. Early goal-directed nutrition versus standard of care in adult intensive care patients: the single-centre, randomised, outcome assessor-blinded EAT-ICU trial. Intensive Care Med. doi:10.1007/s00134-017-4880-3

  5. Singer P, Anbar R, Cohen J, Shapiro H, Shalita-Chesner M, Lev S et al (2011) The tight calorie control study (TICACOS): a prospective, randomized, controlled pilot study of nutritional support in critically ill patients. Intensive Care Med 37(4):601–609

    Article  PubMed  Google Scholar 

  6. Heidegger CP, Berger MM, Graf S, Zingg W, Darmon P, Costanza MC et al (2013) Optimisation of energy provision with supplemental parenteral nutrition in critically ill patients: a randomised controlled clinical trial. Lancet 381(9864):385–393

    Article  PubMed  Google Scholar 

  7. Casaer MP, Mesotten D, Hermans G, Wouters PJ, Schetz M, Meyfroidt G et al (2011) Early versus late parenteral nutrition in critically ill adults. N Engl J Med 365(6):506–517

    CAS  Article  PubMed  Google Scholar 

  8. Doig GS, Simpson F, Bellomo R, Heighes PT, Sweetman EA, Chesher D et al (2015) Intravenous amino acid therapy for kidney function in critically ill patients: a randomized controlled trial. Intensive Care Med 41(7):1197–1208

    CAS  Article  PubMed  Google Scholar 

  9. Gunst J, Vanhorebeek I, Casaer MP, Hermans G, Wouters PJ, Dubois J et al (2013) Impact of early parenteral nutrition on metabolism and kidney injury. J Am Soc Nephrol 24(6):995–1005

    CAS  Article  PubMed  PubMed Central  Google Scholar 

  10. Hermans G, Casaer MP, Clerckx B, Guiza F, Vanhullebusch T, Derde S et al (2013) Effect of tolerating macronutrient deficit on the development of intensive-care unit acquired weakness: a subanalysis of the EPaNIC trial. Lancet Respir Med 1(8):621–629

    Article  PubMed  Google Scholar 

Download references

Author information

Authors and Affiliations


Corresponding author

Correspondence to Michael P. Casaer.

Ethics declarations

Conflicts of interest

The authors have no conflict of interest with respect to the content of this editorial.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Casaer, M.P., Reignier, J. & Doig, G. Optimal guidance for early nutrition therapy in critical illness?. Intensive Care Med 43, 1720–1722 (2017).

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: