The adaptive metabolic changes that occur during starvation were described in detail over 30 years ago and are an excellent example of an integrated physiological response that conserves body protein and maintains normoglycaemia.1 Critically ill patients are often malnourished and then starved for several days whilst undergoing the hormonal, metabolic and inflammatory changes that are commonly referred to as the stress response.2 It seems intuitively obvious, therefore, that feeding these patients during the gross catabolic phase of their illness should be beneficial. For many years, the best route for nutritional support was not clear and it is only in the past decade that there has been general agreement that enteral feeding is the method of choice.3 However, some patients do not tolerate enteral feeding or do not obtain sufficient intake orally or enterally to meet their nutritional needs. Total parenteral nutrition (TPN) may be used in these patients either as a supplementary feed or as the only source of nutrition. Evidence to support the beneficial effects of TPN in critically ill patients has been lacking despite studies going back to the 1970s. A meta-analysis of studies of parenteral nutrition up to 1986 found only 11 trials that were randomized or quasi-randomized, and concluded that perioperative TPN was not justified in unselected major surgery.4 A further meta-analysis of TPN was published in 1998 by Heyland and colleagues with the objective of examining the relationship between TPN and mortality and complication rates in the critically ill.5
The authors initially identified 153 citations from a bibliographic search for the years 19801998, and a further 57 articles were found from a review of reference lists and personal files. They examined only studies that evaluated the use of supplementary TPN in patients already receiving enteral feeds or studies of TPN in patients who were otherwise receiving standard care of oral food and i.v. glucose. Studies comparing TPN with enteral feeding or other forms of TPN were excluded. Only 26 randomized trials involving 2211 patients were suitable for evaluating TPN under these conditions.
The primary outcome was perioperative mortality and the secondary outcome was the rate of major complications. The investigators carefully defined major complications and differentiated them from minor complications, such as wound infection, atelectasis, urinary tract infection and phlebitis. Data on the duration of hospital stay were not aggregated into the analysis because of the variable and infrequent reporting methods. We have previously drawn attention to the problems inherent in attempting to use length of hospital stay as a measure of outcome.6 To try to explain the heterogeneity of the data, the authors developed several hypotheses before undertaking the statistical analysis. They looked at the nutritional status of the patients, the methodological quality of the study, the date of publication and the use of lipids, and compared surgical with critically ill patients.
The results, aggregated from 26 studies, showed no effect of TPN on the mortality:risk ratio, which was 1.03 (95% confidence interval 0.811.31). There was a slight decrease, which was not statistically significant, in major complications in patients receiving TPN in the 22 studies in which these were recorded (risk ratio 0.84, 95% confidence interval 0.641.09). When the a priori hypotheses were examined to try to explain these results, it was found that a significant beneficial effect of TPN on major complications was present in malnourished compared with nourished patients, in studies with a low methodological quality score, in studies published earlier than 1988, and in surgical rather than critically ill patients. The only hypothesis that did not show a significant interaction with the complication rate was the use of lipids. It is notable that there was an increased incidence of major complications in critically ill patients receiving TPN that nearly achieved statistical significance (risk ratio 2.40, 95% confidence interval 0.886.58). In contrast, when these hypotheses were examined using mortality as the end-point, only one remained statistically significant in favour of TPN: the surgical patient compared with the critically ill patient (P=0.03). Furthermore, the mortality rate of critically ill patients receiving TPN was significantly increased (risk ratio 1.78, 95% confidence interval 1.112.85). Only 14 studies reported the effects of TPN on the duration of hospital stay, and in eight studies it was shorter in the control group.
Heyland and colleagues comment that, in spite of their intention to summarize the evidence of the effect of TPN on critically ill patients, the meta-analysis of 26 studies included only six studies of typical intensive care unit (ICU) patients. In particular, there were no studies of medical ICU patients or of patients with sepsis, and only limited assessment of trauma patients. It is common practice to aggregate studies of major surgical and ICU patients on the assumption that the physiological responses to injury and starvation are similar. But the subgroup analysis indicates that, with TPN, the complication rate and mortality rate were significantly greater in critically ill patients than surgical patients. It is no longer appropriate to apply data derived from studies of major surgery to ICU patients.
The results of the meta-analysis do not support the continuing use of TPN in critically ill patients. Indeed, the authors suggest that the only possible use for TPN may be in patients who cannot tolerate enteral nutrition. Since this view was based on a single study of surgical patients, in which TPN was no better than i.v. glucose, its relevance to critically ill patients must be doubtful.7 Proponents of the continuing use of TPN in the ICU must provide evidence of benefit in the face of the unambiguous conclusions of the recent meta-analysis. In our view, TPN in the critically ill patient is of historical interest only, and this form of nutritional support should be confined to the malnourished, surgical patient.
With the popularization of the enteral route for feeding has come the realization that the incorporation of other compounds in the standard enteral feed may have beneficial effects on the metabolic, inflammatory and even immune response to injury. Novel substrates, such as medium-chain and short-chain fatty acids, glutamine and branched-chain amino acids, have been investigated.8 More recently, attention has turned to the use of arginine, glutamine, nucleotides and omega-3 fatty acids as immunomodulators. The notion is simple and attractive: enhance immune function in the critically ill by the use of key nutrients in the enteral feed and the incidence of infectious complications and even the mortality rate should be improved. Immunonutrition was reviewed last in this journal in 1996.9
The most commonly used commercial preparation, Impact, contains the basic amino acid arginine, omega-3 fatty acids and RNA. The evidence to support the immune-enhancing properties of these three compounds is of varying quality. There is a considerable literature on the immunological effects of arginine and also of glutamine, which is incorporated in another commercial preparation, Immun-Aid. These amino acids enhance cellular immunity, modulate tumour cell metabolism, augment lymphocyte and macrophage proliferation (particularly glutamine), improve wound healing and decrease nitrogen loss postoperatively.1013 The ability of glutamine to maintain intestinal barrier function and prevent the translocation of bacteria and endotoxins from the gut lumen to the circulation may be particularly important in critically ill patients.14 There are other properties of arginine that may also be important in modifying the physiological responses to injury. Arginine stimulates the secretion of several hormones, including the anabolic hormone insulin, and prolactin and growth hormone.15 It may also be involved in the local regulation of tissue blood flow as it is a precursor of nitric oxide.16
The omega-3 fatty acids were thought initially to enhance immune defences through increased production of prostaglandin E3 (PGE3), which is less immunosuppressive than PGE2, which is usually released in the inflammatory response. PGE3 down-regulates the production of the proinflammatory cytokines tumour necrosis factor- and interleukin-Iß (IL-Iß), and the major cytokine IL-6. The omega-3 fatty acids also have effects on the release of thromboxane A2 and PGI2 (prostacyclin), and inhibit a variety of cellular and humoral immunological mechanisms. The effects of these compounds on immune function have been reviewed in detail, but the literature pertains mainly to surgical oncology patients.17 18
The administration of RNA or synthetic polynucleotides to enhance immune function is less convincing. Dietary nucleotides are essential for cell-mediated immunity and T-lymphocyte function.19 20 However, most experimental work has been undertaken in animals using nucleotide-free diets over several weeks, and the relevance of much of these data to critically ill patients is obscure. In 1995, Heyland and colleagues reviewed the available studies and concluded that there was little evidence to support the use of dietary RNA supplementation in patients.21 We suggest that this opinion is still valid.
Cerra and colleagues published a preliminary study in 1990 that showed the effects of a combination of nutrients in enhancing immune function in ICU patients.22 This seminal work has stimulated further clinical studies in the past decade to examine not only immune function but also the potential clinical benefits in patients receiving immunonutrition. In 1999, Heys and colleagues reported a meta-analysis of randomized controlled trials of enteral immunonutrition in patients with critical illness and cancer.23 The authors conducted a search for articles published between 1990 and 1998 and identified 11 randomized controlled trials evaluating the use of enteral nutritional support supplemented with combinations of key nutrients (immunonutrition) versus standard enteral nutrition. They evaluated only the clinical outcome: major infectious complications, nosocomial pneumonia alone (as a possible indicator of changes in immune function) and mortality. The duration of stay in the ICU and hospital were analysed when the information was documented. In eight studies, Impact (L-arginine, n-3 essential fatty acids and RNA) was used, Immun-Aid (L-arginine, L-glutamine, branched-chain amino acids, n-3 essential fatty acids and RNA) was used in two studies, and in one study the experimental diet was enriched with L-arginine and L-glutamine.
When the data were aggregated for the meta-analysis, there was a significant decrease in the risk of developing major infectious complications in the immunonutrition group (odds ratio 0.47, 95% confidence interval 0.320.70). However, there was no significant improvement in the incidence of nosocomial pneumonia (odds ratio 0.91, 95% confidence interval 0.531.56). Supplemented enteral nutrition was associated with a significant decrease in the duration of hospital stay of 2.5 days (95% confidence interval 1.04.0), although the duration of stay in the ICU did not change in the four studies in which this variable was noted.
At first sight, the results of the meta-analysis are impressive, with a lower risk of developing a major infectious complication and a shorter duration of hospital stay in the patients given immunonutrition. In contrast, however, there was an increased risk of death in the immunonutrition group (odds ratio 1.77, 95% confidence interval 1.003.12) that just failed to reach statistical significance. It is possible that the lower incidence of major infections and shorter hospital stay in the supplemented group may simply have been a consequence of the increased early mortality. There is no mention in most of the studies analysed of the bias of death on these variables and the need to censor this effect. Indeed, there is an uncanny parallel between the findings of the meta-analysis and a study of immunonutrition in critically ill patients that was published in 1998 after the literature search.24 In this recent investigation of 101 patients, the use of Impact was found to have no effect on mortality (48% in the Impact group, 44% in the control group), but was associated with a significant reduction in the need for pulmonary ventilation (6.0 vs 10.5 days, P = 0.007) and a decrease in the duration of hospital stay (15.5 vs 20 days, P = 0.03). When the duration of hospital stay of patients who died was removed from the results, the difference was no longer statistically significant (P=0.08).
It is important to note that, in the meta-analysis of immunonutrition, the majority of studies (six out of 11) examined surgical patients usually with an upper gastrointestinal malignancy. In four of the surgical studies no patients died. Only five studies investigated critically ill patients, and these were predominantly patients with major trauma. As in the TPN meta-analysis,5 results from typical ICU patients were not available to assess the effects of immunonutrition.
Obviously, there is a need for further randomized controlled studies of immunonutrition in critically ill patients. Future studies must have adequate statistical power, well-described patient groups, precisely defined nutritional regimens, an appropriate control nutritional group and accurate outcome measures. The results of the meta-analysis of immunonutrition hint at beneficial effects: decreases in the incidence of major infectious complications and in the length of hospital stay. It is imperative that appropriate studies of immunonutrition are undertaken to avoid the confusion and uncertainty that has accrued after 25 years of research into TPN. At present, the value of immunonutrition in critically ill patients is not proven.
B. C. Kennedy
G. M. Hall
St George's Hospital Medical School
London SW17 0RE
UK
References
1 Cahill GF, Herrera MG, Morgan AP, et al. Hormonefuel interrelationships during fasting. J Clin Invest 1996; 45: 175169
2 Rolih CA, Ober KP. The endocrine response to critical illness. Med Clin North Am 1995; 79: 2114[ISI][Medline]
3 Heyland DK, Cook DJ, Guyatt GH. Enteral nutrition: a critical appraisal of the evidence. Intensive Care Med 1993; 19: 43542[ISI][Medline]
4 Detsky AS, Baker JP, ORourke K, Goel V. Perioperative parenteral nutrition: a meta-analysis. Ann Intern Med 1987; 107: 195203[ISI][Medline]
5 Heyland DK, MacDonald S, Keefe L, Drover JW. Total parenteral nutrition in the critically ill patient: a meta-analysis. JAMA 1998; 280: 201319
6 Chumbley GM, Hall GM. Recovery after major surgery: does the anaesthetic make any difference? Br J Anaesth 1997; 78; 3479
7 Sandstrom R, Drott C, Hyltander A, et al. The effect of postoperative intravenous feeding (TPN) on outcome following major surgery evaluated in a randomized study. Ann Surg 1993; 217: 18595[ISI][Medline]
8 Grant JP. Nutritional support in critically ill patients. Ann Surg 1994; 220: 6106[ISI][Medline]
9 OLeary MJ, Coakley JH. Nutrition and immunonutrition. Br J Anaesth 1996; 77: 11827
10 ORiordain MG, Fearon KCH, Ross JA, et al. Glutamine-supplemented total parenteral nutrition enhances T-lymphocyte response in surgical patients undergoing colorectal resection. Ann Surg 1994; 220: 21221[ISI][Medline]
11 Park KGM, Heys SD, Blessing K, Eremin O, Garlick PJ. Stimulation of human breast cancers in vivo by dietary L-arginine. Clin Sci 1992; 82: 4137[ISI][Medline]
12 Ardarwi MSM. Glutamine and glucose metabolism in human peripheral lymphocytes. Metabolism 1988; 37: 99103[ISI][Medline]
13 Stehle P, Mertes N, Puchstein C, Zander J, Albers S, Lawin P. Effect of parenteral glutamine peptide supplements on muscle glutamine loss and nitrogen balance after surgery. Lancet 1989; 1: 2313[ISI][Medline]
14 Windmueller HG, Spaeth AE. Uptake and metabolism of plasma glutamine by the small intestine. J Biol Chem 1974; 249: 50709
15 Chang DW, DeSanti L, Demling RH. Anticatabolic and anabolic strategies in critical illness: a review of current treatment modalities. Shock 1998; 10: 15060
16 Gianotti L, Alexander JW, Pyles T, Fukushima R. Arginine-supplemented diets improve survival in gut-derived sepsis and peritonitis by modulating bacterial clearance. The role of nitric oxide. Ann Surg 1993; 217: 64453[ISI][Medline]
17 Heys SD, Park KGM, Garlick PJ, Eremin O. Nutrition and malignancy: implications for surgical practice. Br J Surg 1992; 79: 61423[ISI][Medline]
18 Heys SD, Gough DB, Khan L, Eremin O. Nutritional pharmacology and malignant disease: a therapeutic modality in patients with cancer? Br J Surg 1996; 83: 60819[ISI][Medline]
19 Van Buren CT, Kulkarni AD, Schandle VB, Rudolph FB. The influence of dietary nucleotides on cell-mediated immunity. Transplantation 1983; 40: 6947
20 Rudolf FB, Kulkarni AD, Fanslow WC, Pizzini RP, Kumar S, Van Buren CT. Role of RNA as a dietary source of pyrimidines and purines in immune function. Nutrition 1990; 6: 4552[ISI][Medline]
21 Heyland DK, Cook DJ, Guyatt GH. Does the formulation of enteral feeding products influence the infectious morbidity and mortality rates in the critically ill patient? A critical review of the evidence. Crit Care Med 1995; 22: 1192202[ISI]
22 Cerra FB, Lehman S, Konstantinides N, Konstantinides F, Shronts EP, Holman R. Effect of enteral nutrient on in vitro tests of immune function in ICU patients: a preliminary report. Nutrition 1990; 6: 847[ISI][Medline]
23 Heys SD, Walker LG, Smith I, Eremin O. Enteral nutritional supplementation with key nutrients in patients with critical illness and cancer: a meta-analysis of randomized controlled clinical trials. Ann Surg 1999; 229: 46777[ISI][Medline]
24 Atkinson S, Sieffert E, Bihari D. A prospective, randomized, double-blind, controlled clinical trial of enteral immunonutrition in the critically ill. Guys Hospital Intensive Care Group. Crit Care Med 1998; 26: 116472[ISI][Medline]