Enteral nutrition: better navigation, yet unknown destination?

The nutrition dose truly absorbed by a patient is crucial information in the management or the investigation of nutrition during critical illness. In the present issue of Critical Care, assessment of nutritional losses in stools was studied. These losses together with enteral nutrition lost in gastric fluids and enteral nutrition prescribed but never infused make up the difference between the dose supposedly given to a patient and the amount effectively taken up. Additionally, the optimal dosing and timing of nutrition during critical illness are still debated. When enteral nutrition is insufficient, the options are limited.

In their article in the present issue of Critical Care, Wierdsma and colleagues validated a novel and feasible method to measure the degree to which enteral nutrition (EN) is absorbed by the gastrointestinal tract. Th e accuracy of simply weighing daily faecal production to identify gastrointestinal dysfunction was validated against three reference methods [1]. Energy losses in faeces were measured by laboratory-based bomb calorimetry. Th e contribution of protein and fat to the faecal nutritional losses were estimated by labour-intensive chemical analyses [2,3]. Th e authors identifi ed a daily faecal weight above >350 g as a reliable indicator for gastrointestinal malabsorption. Th e correlation between intestinal energy malabsorption and measured faecal weight was highly signifi cant [1]. Of course, these results need to be confi rmed in a larger study population, including the most critically ill and those with known gastrointestinal problems.
Th e present study is of methodological and conceptual importance to nutritional research and clinical nutrition management. First, the validation of this new assessment technique has been done in a very accurate and complete way and thereby provides a new reliable tool. Secondly, these results focus on a rarely addressed problem in the critically ill: is the EN administered to a patient truly absorbed? In 9 out of the 48 stable patients in this trial, the nutrition was only partially absorbed. Th is 19% represents a high incidence of gastrointestinal dysfunction since patients with known gastrointestinal problems were not included [1]. Previous studies assessed and treated diarrhoea in critically ill patients, measuring the frequency, liquidity and volume of stools but not the proportion of EN energy, proteins and fat lost by the patient [4,5].
Whether studying the eff ect of nutritional interventions or managing nutrition in clinical practice, we will have to take into account these data on gastrointestinal energy losses. It is also timely to remind researchers and clinicians of other confounders. Th e discrepancy between prescribed and eff ectively infused EN doses in critically ill patients is relevant. Interruptions of EN for procedures, emesis, high gastric residual volumes (GRVs), diarrhoea and feeding tube replacement are important culprits [6]. EN lost in GRVs being discarded after GRV assessment can be measured using a refractometer [7]. An alternative is to reinject GRVs after measuring or to refrain from measuring GRVs, since evidence supporting this practice is limited [8]. Briefl y, we should be aware that the amount of EN eff ectively taken up by the patient is the prescribed volume minus the volume not administered to the patients, minus the EN discarded with GRV minus the EN lost in faeces.
Two more questions remain, however. How much EN do we want to give? And what can we do when EN does not achieve this goal?
First, how should we determine the nutritional target? Indirect calorimetry measurement of energy expenditure is an attractive individualised method. Indirect calorimetry, however, does not measure energy needs but energy consumption, and indirect calorimetry is less reliable in patients on renal replacement therapy, aggressive ventilation and other conditions often present in the

Abstract
The nutrition dose truly absorbed by a patient is crucial information in the management or the investigation of nutrition during critical illness. In the present issue of Critical Care, assessment of nutritional losses in stools was studied. These losses together with enteral nutrition lost in gastric fl uids and enteral nutrition prescribed but never infused make up the diff erence between the dose supposedly given to a patient and the amount eff ectively taken up. Additionally, the optimal dosing and timing of nutrition during critical illness are still debated. When enteral nutrition is insuffi cient, the options are limited. critically ill [9]. A recent randomised trial showed more morbidity in the ICU, but an unexplained reduction in hospital mortality in the per-protocol analysis, when feeding was guided by indirect calorimetry [10]. Th e next issue is how fast the nutritional target should be reached. Recent trials found no benefi t in early EN administration up to the calculated target as compared with low EN intake [11,12].
Finally, what should be done if EN uptake is insufficient? Probiotics or fi bre-rich EN reduced the volume or frequency of diarrhoea in critically ill patients in some of the few randomised controlled trials on this topic [4,5]. Whether this reduction also improved gastrointestinal absorption is unknown. Absorption of EN by dys functional gut could be facilitated by the absence of proteins or even peptides in, respectively, semi-elemental EN or elemental EN; how ever, trials in critically ill patients failed to provide convincing evidence [13,14]. Th e best controlled method to eff ectively complete insuffi cient EN would be the intravenous administration of nutrition: parenteral nutrition. Supple ment ing insuffi cient EN with parenteral nutrition during the fi rst week of critical illness, however, resulted in more infections and delayed recovery in a large randomised controlled trial [15].
In conclusion, the optimal EN dose during diff erent stages of critical illness is not yet known. Stool weight measurements could improve our knowledge of true EN absorption, however, and could thus lead to more accurate research and clinical practice.

Competing interests
KU Leuven received an institutional partial (<30%), unconditional and nonrestrictive research grant for the EPaNIC trial from Baxter Healthcare (Maurepas, France). The authors declare that they have no competing interests.