In this issue of Intensive Care Medicine, Keift et al. [1]report the results of a multicenter, randomized, double-blind, controlled trial comparing an enteral formula en-riched with arginine, glutamine, antioxidants and con-taining omega 3 fatty acids with an isocaloric, non-isonitrogenous standard enteral formula, in a mixedpopulation of 597 critically ill patients. When analyzedboth using an efficacy analysis (those fed for more than48 h) and by an intention-to-treat analysis, the authorsfailed to demonstrate any difference in clinical outcomesin the overall analysis or several subgroup analyses.Mortality, infectious complications, ICU length of stay,and duration of mechanical ventilation were the samebetween the two groups. Although the lack of a treatmenteffect on clinical outcomes has been shown in other trials,given the highly controversial nature of the topic of im-munonutrition in critically ill patients, this article is sig-nificant, particularly since this is the largest and one ofthe more robust randomized trials of immune-enhancingdiets in an ICU population.When the results of this large trial are added to theexisting trials, in the form of a meta-analysis, the overalltreatment effect is consistent with no effect on mortality(relative risk [RR] 1.05 and 95% confidence intervals [CI]0.89, 1.25), infectious complications (RR 0.97, CI0.81,1.15) or length of stay in ICU (weighted mean dif-ference [WMD] 1.59, CI 3.25,0.08) [2]. After 20 yearsof research, 18 randomized trials of 2,348 critically illpatients, and countless millions of research dollars, why isit that we are unable to demonstrate a positive treatmenteffect associated with these immune-enhancing products?Do we need more trials like this one to prove the virtuesof these special nutrients? Or, do we have to ask ourselveswhether there are fundamental problems with our ap-proach to proving the hypothesis that certain nutrientswith immunologic and metabolic properties will improvethe outcomes of critically ill patients? We have arguedbefore that there are significant methodological limita-tions to the existing literature that limit the inferences wecan make from these studies [3]. As we argue below,perhaps the problem is further compounded by the factthat multiple nutrients are combined into single nutri-tional products and tested in heterogeneous patient pop-ulations.For any sick patient, “metabolic resuscitation” of thegastrointestinal tract by providing adequate nutrition, ingeneral, and defined immuno-modulating substratesspecifically, to maintain gut-barrier integrity and functionand to reduce regional oxidative stress, will have to beconsidered as key therapeutic strategies. However, withina given patient over time or across different patientpopulations, the severity of “gut failure”, the amount ofbacterial translocation, the degree of cellular immunedysfunction, the balance of inflammation/anti-inflamma-tion, and the regional and systemic generation of reactiveoxygen species will vary. Generally speaking, electivesurgical patients experience minimal activation of cy-tokines and some degree of suppression of the cellulardefense function following surgical stress, putting them athigher risk for acquired infectious morbidity and mor-tality. It follows that nutrients, such as arginine, thatstimulate the cellular defense system may reduce infec-tious complications in the elective surgical patient. Incontrast, the associated changes to the systemic inflam-matory response accompanying critical illness are farmore intense, complex, variable, and less well-defined,but best characterized by an over-amplified inflammatory
Read full abstract