Metabolic support in the ICU continues to advance and one overt by-product is the need for all physicians on the team to be educated about relevant innovations. This issue of Nutrition and the ICU covers a broad range of nutritional topics from different nations. A common thread is that many organ systems and new technologies are involved. The reader may sense that the practical assimilation of new information must be objective and algorithms for the formulation of metabolic support care plans evolve over time and are therefore ‘moving targets’.
Gastrointestinal disorders remain a major obstacle to the more widespread use of enteral nutrition. This problem is compounded by our inability to accurately assess this invisible organ. None of the available physiology or organ failure scores includes yet any gut failure assessment: scoring attempts have failed and prompt further studies . In this issue, Röhm et al. (pp. 161–167) provide an excellent review of motility disorders with appropriate treatments and they emphasize the importance of having efficient prokinetics to implement early enteral nutrition support. Cisapride, a 5-hydroxytryptamine 4 [5-HT4]) (serotonin) receptor agonist, was a very efficient propulsive agent, which was withdrawn from the market in 1996 due to case reports about fatal torsades de pointe. Tegaserod another relatively new 5-HT4 agonist with moderate affinity for the 5-HT1-receptor, was withdrawn recently due to an increased rate of myocardial infarction. The more widely used drugs metoclopramide and domperidone are not very efficient, acting only on the upper gastrointestinal tract, whereas the relatively older drug, neostigmine, remains quite efficient in promoting colonic propulsion. Some hope comes from new agents under investigation such as cholecystokinin -antagonists, motilin-like agents, other 5-HT5 agonists, and mu-receptor antagonists.
The exocrine pancreas is another hidden organ. Hardt et al. (pp. 168–174) present a cogent argument that the ‘pancreatitis of critical illness’ is multifactorial in nature and is, in fact, much more common than most physicians have suspected, affecting up to 80% of ICU patients. Intermediation of pancreatic function among injury and inflammation of critical illness, glycemic control and clinical outcome is a potential therapeutic target in the ICU. Notwithstanding the evidence-based controversies surrounding enteral nutrition versus parenteral nutrition, early nutrition support, and gastric versus postpyloric enteral nutrition, various nutritional pharmacological interventions hold promise. These include the use of probiotics, n-3 fatty acids, glutamine, and certain fibers.
Some years back the Lausanne ICU team identified energy balance and energy debt as an important contributor to ICU-related complications and outcome . They also showed, using a computerized monitoring system, that they were not only delivering the patients with energy from feeds but also significant amounts of incidental energy with dextrose used for drug dilution and propofol lipid emulsions . This finding was made in the context of preventing the complications directly resulting from ‘underfeeding’. Are we now pushing the pendulum too far, going from underfeeding to overfeeding? With their paper on the impact of glucose and fat on liver function, Grau and Bonet (pp. 175–179) propose an analysis of energy delivery in the ICU, which transcends the pure concept of calories. In their 2007 paper, Grau et al. had identified an energy intake cut off of 25 kcal/kg above which the risk of developing liver test alterations increased steeply. They now show that it is not only the total amount of calories that matters but also the distribution of calories between glucose, fat and proteins . This confirms some recent data that were presented at the 2008 ESPEN (European Society for Clinical Nutrition and Metabolism) congress in Firenze . In a database analysis including 1209 patients, early energy supply (first 3 days) more than 1500 kcal distinct from parenteral glucose is associated with lower mortality and morbidity in ICU patients . Given the pervasive confusion on optimal energy requirements in the ICU and the failure of predictive equations and indirect calorimetry to provide a universally accepted answer, evidence-based results that hone in on a generalizable target range of calories is most welcome.
Differences in gastrointestinal complications between adults and children are analyzed by Lopez-Herce (pp. 180–185). Although children share some characteristics with critically ill adults, they are not small adults and the incidence of complications in children appears to be lower than in adults. In pediatrics, the most important risk factors for digestive tract complications are shock and the administration of drugs (catecholamines, sedatives, and muscle relaxants). But in adults, altered gastrointestinal motility is the principal mechanism underlying large gastric residues, abdominal distension, and constipation. It is not yet clear to what extent evidence pertaining to gastrointestinal complications of nutrition support in the adult ICU patient can be extrapolated to pediatrics.
Only a few centers have experience with enteral nutrition after intestinal transplantation. Colomb (pp. 186–189) discusses the rationale for using early dilute semielemental enteral nutrition in the setting of impaired lymphatic circulation of the graft. The absence of consistent enteral nutrition protocols among the various intestinal transplant programs signals the needs for more investigative data and its interpretation by expert multidisciplinary teams.
One under-recognized complication of critical illness, amenable to specialized management, is metabolic bone disease. This potentially devastating condition, whose phenotype is typically expressed in the post-ICU or even posthospitalization setting, is bone fracture and chronic debilitation. Hollander and Mechanick (pp. 190–195) explore the emerging database on the use of bisphosphonate therapy in critical illness. Although most data are restricted to the ‘chronic critical illness’ setting, one wonders whether early, preemptive use of bisphosphonates, perhaps with vitamin D analogues, might prevent the development of subsequent bone and mineral abnormalities that could affect clinical outcome.
The theme of technological advances in the ICU is explored by Scurlock et al. (pp. 196–200). Two interventions gaining more widespread use in the cardiothoracic ICU are various renal replacement therapies and ventricular assist devices. Both interventions are utilized in patients who are also at high nutritional risk and their implementation exacerbates this risk. The question at present is whether in the absence of strong clinical outcome data, there is a role for ‘intensive metabolic support’ , that is, early enteral nutrition ± combined parenteral nutrition in these patients?
Finally, evidence-based medicine (EBM) is a hard rule. Each of the above papers from this issue suggests the need for more studies to clarify a difficult clinical nutrition problem. However, many attempts at evidence-based clinical investigation contain flaws. Gaps between clinical evidence and clinical practice are less likely to occur when strong evidence demonstrates clear patient-oriented benefits . Another pitfall in EBM is the application of ‘formal rules of evidence’ to determine the validity of a research paper before deciding whether it should be used to guide clinical decisions . The paper by Doig et al. (pp. 201–206) exposes other shortcomings of published EBM papers; there is another gap between research that was properly carried out and the published form. Reporting details, such as an intent-to-treat analysis, allocation concealment, and blinding are areas for improvement. The meta-analysis methodology is also called into question when the result is better than the original data. Publishers, journal editors, and reviewers share in the responsibility for an unbiased and balanced representation of research findings in the literature. This is difficult when the majority of journals opt out of publishing negative randomized controlled trials based on ‘impact factor’ concerns. Nevertheless, the quality of clinical trials appears to be improving and this paper emphasizes the importance of a coherent global quality procedure.
What is the take-home message for this issue? New ideas, technologies, and approaches are developed from different starting points, progress along different paths at different paces, and incorporate evidence bases that may be unique to a particular geography. Improving our implementation of nutrition support in the ICU depends not only on experience-based learning, but also reading, conferencing, and other forms of discourse among many medical communities. It is hoped that optimization of EBM will incite optimization of clinical nutrition and metabolic care in the ICU.
1 Berger MM, Oddo M, Lavanchy J, Longchamp C, Delodder F, Schaller MD. Gastrointestinal failure score in critically ill patients (Letter). Crit Care 2008; 12:436.
2 Villet S, Chioléro RL, Bollmann MD, et al
. Negative impact of hypocaloric feeding and energy balance on clinical outcome in ICU patients. Clin Nutr 2005; 24:502–509.
3 Berger MM, Revelly JP, Wasserfallen JB, et al
. Impact of a computerized information system on quality of nutritional support in the ICU. Nutrition 2006; 22:221–229.
4 Grau T, Bonet A, Rubio M, et al
. Liver dysfunction associated with artificial nutrition in critically ill patients. Crit Care 2007; 11:R10.
5 Pichard C, Kreymann GK, Weimann A, et al
. Early energy supply decreases ICU and hospital mortality: a multicenter study in a cohort of 1209 patients. Clin Nutr 2008; 3(Suppl 1):7 (O015).
6 Mechanick JI, Chiolero R. Special commentary: a call for intensive metabolic support. Curr Opin Clin Nutr Metab Care 2008; 11:666–670.
7 Grilli R, Lomas J. Evaluating the message: the relationship between compliance rate and the subject of a practice guideline. Med Care 1994; 32:202–213.
8 Evidence-Based Medicine Working Group. Evidence-based medicine. A new approach to teaching the practice of medicine. JAMA 1992; 268:2420–2425.