Article Text

Feeding the inflamed pancreas
  1. Department of Gastroenterology,
  2. Ninewells Hospital and Medical School,
  3. Dundee DD1 9SY, UK

    Statistics from

    See article on page 431

    There is a renewed awareness of the detrimental impact of undernutrition on organ function and clinical outcome. There is also evidence that in some clinical circumstances nutritional support will reverse nutritional depletion and improve recovery.1Nutritional recovery may not be achieved in the metabolically stressed patient.2 The effects of nutritional depletion occur early,3 and subsequent recovery may be prolonged in the severely depleted patient.4 Thus, there is emphasis on early nutritional intervention to prevent or retard the development of malnutrition in patients who are unable to eat or absorb an adequate oral diet. The British Society of Gastroenterology guidelines suggest that such patients should not be deprived of nutrition for more than seven days, and that nutritional support should be considered earlier in patients who are already malnourished or who are metabolically stressed.5 The conventional management of patients with acute pancreatitis includes parenteral nutrition, which is instituted early in the course of the illness, partly because of the frequently associated ileus, and partly for fear of the consequences of stimulating pancreatic function by oral nutrition or enteral tube feeding.

    Parenteral nutrition is not only expensive, it is potentially hazardous. Problems include complications of catheter placement, catheter associated infection and central vein thrombosis. Although parenteral nutrition will provide most essential nutrients, conditionally essential nutrients such as glutamine are not included in conventional solutions. This may be one reason why parenteral nutrition does not protect gut barrier function, at least in animal experiments. There is an abundance of animal data that luminal nutrition is important for this aspect of intestinal function.6 7Without luminal nutrition, or possibly in the absence of specific nutrients such as glutamine, there is an increase in intestinal permeability to toxins, and increased translocation of bacteria.7 Thus, in the critically ill patient the gut has been considered the motor of multisystem failure with increased permeability and translocation accentuating the cytokine response. Human data are sparse. Some authorities have reported translocation as a spontaneous event which is not increased by bowel rest and conventional parenteral nutrition in the stable patient.8Recent data suggest that enteral nutrition is feasible in patients with pancreatitis.9 Thus, a comparison of enteral and parenteral feeding in patients with acute pancreatitis is interesting and important.Shouval

    The study by Windsor and colleagues in this issue (see page 431) compares total parenteral nutrition (TPN) with total enteral nutrition (TEN) in patients with acute pancreatitis. Patients with mild or moderate disease were given oral supplements to tolerance or parenteral nutrition through a peripheral vein. Patients with severe disease received nasojejunal feeding or central parenteral nutrition. The groups were matched for disease severity. Seven days after beginning nutritional support there was an improvement in the APACHE score and C reactive protein, and reduced length of stay in the ITU in the enterally fed patients in comparison with the patients who received TPN. Furthermore, there was an increase in the anti-endotoxin antibody values and a decrease in the total antioxidant capacity in the TPN group, the antibodies did not change and the antioxidant capacity increased in the TEN group.

    The apparent benefit of enteral feeding compared with parenteral feeding is in keeping with current views on the importance of luminal nutrition for the preservation of the intestinal barrier function. However, there are alternative explanations for the differences between the groups in this study. These relate to the amount of nutrition provided and the nature of the energy substrates.

    The median number of non-protein calories delivered to the enteral group was 1201 kilocalories per patient day. Oral supplements were given to tolerance and temporary reduction in the volume of feed was required in five of the 16 patients randomised to enteral feeding. By contrast, the parenteral group received all 1800 non-protein calories. We know that the catabolic response to stress cannot be switched off by nutritional support. Cytokines induce proteolysis and lipolysis which provide endogenous substrate for the inflammatory response. Under these circumstances exogenous carbohydrate and lipid may lead to hyperglycaemia and hyperlipidaemia. Thus, nutrition support may be harmful in patients with more severe disease. The possibility that the enteral group fared better because of hypocaloric feeding, or that nutrition support is disadvantageous in this context, cannot be excluded in the absence of a control group maintained on electrolyte infusions for the first week of illness.

    Patients who were randomised to receive parenteral nutrition were given 9.4 g nitrogen, and 1800 non-protein kilocalories, of which 990 calories were supplied as long chain triglycerides. Conversely, the enteral groups were given nutrient solutions in which approximately one third of the energy was supplied as fat. Given the reduced enteral nutrient delivery, this would amount to approximately 40% of the fat supplied to the parenteral group. The influence of this difference on the modulation of immune function and the generation of proinflammatory cytokines in this context is speculative and uncertain. Nevertheless, long chain triglycerides of the type that are administered during conventional parenteral nutrition have metabolic and immunological effects particularly in the stressed patient. They increase the production of metabolites of arachidonic acid and they may suppress mononuclear phagocyte function.10

    This is an important clinical study from which we can draw two conclusions. Some of the nutritional needs can be safely provided by the enteral route in patients with acute pancreatitis. Patients who receive enteral feeding fare better than those who are given parenteral nutrition. The reasons for these findings merit further investigation.

    See article on page 431


    View Abstract

    Request permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Linked Articles