A critically ill patient will preferentially catabolize lean body mass when they are not provided with sufficient calories. This loss of lean body mass negatively impacts immune function and wound healing, and may also affect survival. The enteral route is understandably the safest, most convenient, and most physiologically sound method of nutritional support. However, most critically ill animals cannot readily tolerate enteral support due to gastrointestinal dysfunction. In such cases, parenteral nutrition (PN) should be considered.

PN is a mixture of amino acids, glucose and lipids, and each of these components has effects that may compromise critically ill patients. Simply stated, PN is not without complications and devising a formulation that optimizes recovery and minimizes complications is currently challenging (Table 1).

Table 1. Common Complications of Parenteral Nutrition.

PN has a reputation of causing more harm than good. A recent study showed early total parenteral nutrition (TPN) in more severely ill veterinary patients was associated with lower survival rates. Some believe many of the misconceptions and complications may be due to over-estimation of caloric needs rather than the form of nutritional support. Furthermore, gut atrophy and gut-associated sepsis with PN administration occurs because there is a lack of enteral nutrients, not because nutrients are being administered intravenously.

Lipids

Lipid emulsions are traditionally composed of artificial chylomicrons and are stabilized with glycerol phospholipids. They are typically derived from soybean or safflower oil, and the triglycerides are long-chain polyunsaturated fatty acids. Lipid-derived fatty acids provide essential building blocks for cell membranes and prostaglandin metabolism. However, parenteral lipid may pose risks given the risk for infections with the inflammatory, catabolic state of critical illness. Table 2 summarizes some of the major benefits and risks of parenteral lipid therapy.

Table 2. Characteristics of Lipid Emulsions.

Several studies evaluating lipid-free PN in human populations have documented lower incidences of infections in critically ill patients. Although no difference in survival has yet been reported, hospital length of stay, length of time spent in ICU, and ventilator-free days were less when lipid-free PN was used. Furthermore T-cell function was improved 5 days after initial injury with lipid-free PN, while immune function worsened in individuals receiving lipid-containing PN.

There are no studies specifically addressing the role of lipids in parenteral nutrition for critically ill animals. The vast majority of publications detailing PN use in animals have included lipid in the formulation and as such, there are no controlled trials evaluating how lipid inclusion impacts complication rates.

Hyperglycemia

Hyperglycemia, in critically ill humans, is a poor prognostic factor associated with several conditions. A pilot study, evaluating the incidence of hyperglycemia in cats presented as emergencies, found hyperglycemia was associated with poorer outcome. Another study evaluating hyperglycemia in critically ill dogs, found mortality rates dramatically increased as patients become progressively more hyperglycemic. The presence of hyperglycemia in dogs and cats with traumatic brain injury was correlated with severity of injury, but there was no significant relationship between the degree of hyperglycemia and outcome.

Reducing hyperglycemia with insulin therapy has yielded improved outcomes in several human populations. A study of cats receiving TPN, documented development of hyperglycemia at 24 hours was associated with poor outcome. There is no meaningful evidence in animals that treating this hyperglycemia is beneficial. Nevertheless, limiting risk factors for the development of hyperglycemia is appropriate. Although guidelines for controlling hyperglycemia in veterinary patients receiving PN have not been devised, the use of exogenous insulin in these patients may be appropriate. Prospective research on this issue is needed.

Amino Acids

Mobilization of the amino acid (AA) pool from muscle tissue, is a principal response to injury and inflammation. Specific AA deficiencies may result from utilization in excess of the amount obtained from the diet (Table 3). As such, parenteral administration of AA solutions with modified ratios of branched-chained to aromatic amino acids, were marketed as an important strategy in critically ill patients. At this time there has been no documented clinical benefit of these solutions.

Table 3. Common Amino Acid Deficiencies in Critically-Ill Dogs.

 Alanine

 Citrulline

 Glycine

 Methionine

 Proline

 Serine

 Arginine

 Glutamine

Glutamine is the preferred fuel source for enterocytes and cells of the immune system. In critical illness, depletion of this amino acid may compromise immune function and contribute to gut-derived sepsis. In humans, enteral glutamine therapy has been shown to have some positive results. Additionally, parenteral provision of glutamine, as a dipeptide, also produced some very encouraging results in human populations. Parenteral glutamine has not yet been evaluated in veterinary patients.

Arginine is an extremely important amino acid for both the maintenance of immune function and as a precursor for nitric oxide (NO). With up-regulation of NO synthase during critical illness, arginine stores become rapidly depleted. Interestingly, arginine supplementation may potentially compromise patients given NO's vasodilatory effects, and prospective research is needed to investigate this issue.

When calculating the number of calories required for a patient, some veterinary nutritionists aim to meet all caloric requirements with lipids and dextrose; they then provide additional AA as required by the patient (usually 4–5g/100 kcal) but do not take into account the calories this component provides (~4 kilocalories per gram once metabolized). The belief is an animal not provided with their entire energy requirements in the form of dextrose and lipids will use the provided AA for energy rather than protein synthesis. However, at this time, there is little evidence to support this hypothesis. Overall, there is evidence to show provision of more calories than the estimated RER is associated with the development of hyperglycemia and possibly decreased survival. In order to avoid over-feeding, the author does account for the energy provided by AA, when formulating PN to minimize the incidence of hyperglycemia.

References

References are available upon request.