Lipid Emulsions:

Dosage and Monitoring

Olivier Goulet, M.D.

Gastroentérologie et Nutrition, Hôpital Necker-Enfants Malades, 75743 Paris cedex 15, France

phone +33-1-4449-4412, fax +33-1-4449-2501, email:

Learning Objectives

  • To understand the optimal glucose / fat ratio
  • To be familiar with methods for adequate assessment of the intravenous lipid clearance and tolerance
  • To identify the mechanisms of lipid overload and the patients at risk
  • To understand possible relation between long-term fat infusion and liver disease

Introduction

After a period of development of the technique and of improvement in prevention of complications, parenteral nutrition (PN) is now widely used in a variety of indications in adult as well as in pediatric patients. Macro- and micronutrients given exclusively through the central vein for prolonged periods are sufficient for nutritional recovery and maintenance as well as growth in children.

One of the current objectives is to adapt the PN intakes to the clinical and nutritional situation. Indeed, it is essential that the composition of the intravenous feeding solution be adjusted according to age and disease, to the clinical and biological criteria and as a function of the progressive transition from parenteral to enteral feeding.

Energy intake is closely correlated with that of nitrogen (nitrogen/energy ratio). Glucose is the unique carbohydrate used during TPN. The use of other carbohydrates (fructose) is not advised. The infusion rate of glucose must be kept constant without exceeding rate of 1 g/kg per hour or 1.5 g/kg per hour especially in infants and children. Under these conditions the appearance of a glucosuria indicates a technical problem with the infusion rate or some metabolic disorder, especially during sepsis or any situation of stress. Intravenous fat emulsions (IVFE) are usually added to the PN regimen because they provide essential fatty acids and a concentrated source of calories with a relatively low osmotic load. IVFE enables the physician to provide a highly concentrated source: 2.2 kcal/ml for the 20% emulsion, compared to 0.68 kcal/ml for a 20% dextrose solution. The concentrated calories and low osmolality of IVFE make it ideal for peripheral parenteral nutrition. The use of IVFE also allows the energy input to be diversified, with a reduction in the consequences of excessive glucose supply. In addition it was well demonstrated in adults as well as in pediatric patients that the use of IVFE improves net nitrogen balance [1-4].

Optimal Glucose : Fat Ratio

In the past, TPN for adults, children and infants provided most of the energy as glucose, although it was not precisely known how much of the ivadministered glucose was oxidized. Glucosebased TPN has been shown to cause adverse effects related to glucose storage, particularly as fat. This might account for the extensive lipid deposition reported both in liver and adipose tissue [5-6]. In these conditions, fat infusion further increases fat deposition and may result in fat overloading. Thus, substitution of part of the glucose calories avoids the undesirable effects reported with glucosebased TPN. Recent studies performed in infants or neonates have assessed glucose and fat utilization [7]. By using five isocaloric TPN regimes differing in their glucose/lipid ratio, it was possible to assess for the optimal glucose infusion rate [7]. Fat infusion aiming at a significant contribution to the coverage of energy expenditure requires that glucose oxidation be equal to or lower than maximal oxidative glucose disposal.

A study in malnourished infants and young children has shown that the amount of infused lipid must be adapted to lipid oxidation capacity [8]. There is a maximal lipid utilization rate of about 3.3 to 3.6 g/kg/day (infants). Pierro showed similar results in a shortterm study performed in surgical neonates on TPN [9].

Finally, shortterm as well as long-term lipid use must be included in all TPN programs in infants, children and adults.

Assessment of Lipid Clearance

The composition of IVFE results in a mixture of artificial chylomicrons having physicochemical characteristics very similar to those of the natural chylomicrons produced by enterocytes. They are hydrolysed by lipoprotein lipase (LPL) in capillaries of adipose tissue and muscle. This step determines the rate at which fatty acids from the emulsion are transported into the tissue for storage or oxidation. Like many other key enzymes, LPL can be regulated both in amount and in activity. The amount of LPL at the capillary endothelium is under hormonal control, while its activity is regulated by the specific activator apo CII [10].

Lipid clearance test have been developed in the past but are not used routinely in clinical practice. They allow understanding of the factors determining the elimination of exogenous fat emulsions from the circulation [11]. An intravenous fat tolerance test using Intralipid was developed in the late 1960s demonstrating that the emulsion was handled in a similar manner to that of endogenous chylomicrons. Patients with moderate trauma, for example after cholecystectomy, have an increased fractional elimination rate in comparison with healthy controls. Critically ill patients on the other hand have an elimination rate which is lower or close to that of healthy controls [12].

Lipids being not water-soluble have to be transported bound to proteins. The lipoproteins play a key regulatory role in the metabolism and elimination of exogenous lipids. The endogenous VLDL particles undergo degradation to LDL in plasma under the action of LPL. These particles contain apoCII which acts as an activator of LPL. Exogenous TG particles are handled in the same way but have to acquire apo CII mainly from the HDL fraction before they can be hydrolyzed. During the degradation of VLDL particles there is a continuous exchange of apolipoproteins and surface material. Apo C, apo E and phospholipids are removed during lipolysis and transferred to HDL. As donor of apo CII and acceptor of abundant surface material, HDL play a key regulatory role in the metabolism of VLDL and exogenous lipids. As might be expected, the serum HDL concentration is positively correlated to the fractional elimination rate of exogenous fat emulsions. HDL-cholesterol has been shown to be low in critically ill patients as well as Apolipoprotein CII. The low HDL and impaired apolipoprotein pattern taken together is a possible explanation for the disturbance in the elimination capacity of exogenous lipids seen in some critically ill patients.

Patients treated with heparin due to deep vein thrombosis have been studied using the intravenous fat tolerance test. The fractional elimination rate decreased during heparin treatment and after interruption of heparin infusion the elimination rate slowly returned to normal. Heparin acts by activating and releasing LPL from the endothelial surface and when given together with exogenous fat emulsions causes a sharp increase in circulating LPL activity and FFA levels. Sometimes this might be a positive effect, but in patients with already high FFA levels an increase in lipolysis may lead to extremely high FFA values. Heparin is necessary for clinical reasons in many situations and it would be an advantage if the increase in lipolytic activity could be avoided. Low molecular heparin is an interesting alternative in these situations, since its effect on lipolysis is much less than that of unfractionated heparin. There is no place for routine administration of heparin together with the fat emulsion as proposed in the past.

Metabolic monitoring of patients prior to and during infusion of IVFE should include

  • fasting plasma triglycerides (TG)
  • total cholesterol
  • plasma LDL- and HDL-cholesterol.

Normal plasma TG levels do not mean that exogenous triglycerides are adequately used (oxidation) or stored (adipose tissue). Indeed, part of the exogenous lipids may be cleared by other mechanisms, especially removal by the reticuloendothelial system (RES). High plasma TG levels, however, suggest impaired clearance related to excessive infusion rate and/or decreased LPL activity.

There may also be several classical contraindications to the use of IVFE during the initial phase, such as sepsis, thrombocytopenia, disseminated intravascular coagulation, respiratory distress syndrome or metabolic acidosis. Furthermore, administration of IVFE during neonatal jaundice has to be careful as there is a risk of displacing non-conjugated bilirubin from albumin with free fatty acids.

Fat Overload and Liver Disease

The fat overload syndrome (FOS) has been reported a long time ago and was thought to be related to the captation of exogenous particles by the reticuloendothelial system (RES) [13]. Its acute clinical and biological expression was described as mimicking septic syndrome but related to macrophage activation. Symptoms include high fever, hepato-splenomegaly, jaundice, respiratory distress syndrome, bleeding, thrombocytopenia, disseminated intravascular coagulation, metabolic acidosis, and hypoalbuminemia [14]. Cessation of lipid administration most of the time is insufficient for improving the patient’s condition. Steroids have been shown to stop the process of macrophage activation.

More recently we have reported several cases of severe cholestasis supposed to be related to long-term administration of IV lipid [15]. Liver disease is a major side-effect of long-term PN. Fifteen years after the first reports its pathogenesis is not completely understood [16]. Disruption of th enterohepatic circulation of bile acids in case of ileal resection, impairment of choleresis in the absence of oral feeding, bacterial overgrowth due to bowel obstruction, stasis and loss of the ileocaecal valve are patient-dependent factors thought to contribute to PN-associated cholestasis [17]. One mechanism might be an increase in bile concentration of lithocholic acid [18]. Bacterial infections were also proposed as a co-factor in PN-related cholestasis, since the sepsis-associated cholestasis has been well described, both in human and animal models [19-21]. Duration of PN is also a known risk factor. Qualitative aspects of PN are also a matter of debate. Experimental evidence suggested that an excess in total energy delivered induces liver lesions, reversible when decreasing the energy supply [22]. The role of amino acids was suspected, either excess or lack of them [23,24] , and an excessive glucose supply might induce steatosis through the increase in de novo lipogenesis [25].

Although lipids are necessary in patients on PN, they are strongly suspected to be toxic in some cases. The metabolism of their oxidized fraction is relatively well known, far less the destiny of the non oxidized fraction, which is caught by the reticulo-endothelial system in the liver and also by the spleen, bone marrow and lungs. Chronic administration of lipid emulsions might overload reticuloendothelial cells and induce their acute or chronic activation. It must be noticed though that an excess of liver macrophages and portal infiltration with eosinophils has been previously described in PN-dependent patients, but not attributed to lipid toxicity [26]. Accumulation of exogenous lipids in the liver Kupffer cells may also impair the clearance of endotoxins and increase their deleterious effect on the liver. Moreover, the peroxidation of exogenous lipid could produce toxic metabolites despite the simultaneous infusion of vitamin E [27]. Phytosterols contained in lipid emulsions may also have a deleterious effect on biliary secretion [28].

From our recent study, several arguments are in favour of a toxic role of parenteral lipids for the liver. First, it was observed that several episodes of cholestasis occurred after a recent change in the mode of lipid delivery. Second, hematologic disorders compatible with a macrophage fat overload occurred in several cases contemporary to the onset of cholestasis and frequently improved together with cholestasis after suspension or decrease in lipid infusions.

Withholding lipid emulsions from PN-dependent patients is not free from risk. As expected, prolonged lipid withholding induced growth retardation in childhood and essential fatty acid deficiency. The increase in daily glucose supply is not a satisfactory prevention of failure to thrive, because of the induced insulin resistance [29], and the risk of subsequent liver steatosis.

Lipid-induced complications have to be taken into account in the decision for small bowel transplantation (SBT) in children with irreversible digestive disease [30]. Failure to thrive secondary to repeated or prolonged lipid suspension might be an indication for isolated SBT, before the development of severe liver disease. The liver status should be checked regularly by plasma liver tests and liver biopsies in patients on waiting list for isolated SBT, especially in case of mild liver fibrosis at the time of decision for isolated SBT. We have recently showed in a group of children on total PN that severe liver fibrosis may appear over the first 4 years of life and then rapidly progress from septal fibrosis to cirrhosis [31]. Patients who are sensitive to lipid toxicity might be at risk for rapid course of liver fibrosis. Taking into account the waiting delay for transplantation, one could suggest to plan rapidly combined transplantation in children presenting with lipid intolerance and severe liver fibrosis.

Rules of lipid administration in PN-dependent children may be proposed:

  • maximal daily amount of 2 to 2.5 g/kg/d
  • maximal infusion rate of 150 mg/kg/hour
  • no more than 5 lipid infusions weekly
  • maximal lipid-to-energy ratio of 25%.

In children receiving lipid emulsions, liver tests and platelet count should be monitored regularly.

When cholestasis occurs, biliary obstruction, infection or drug toxicity should be ruled out by appropriate investigations. Platelet count <150.000/mm3 associated with an increase in plasma transaminases, a fortiori an increase in plasma bilirubin, is strongly suggestive of lipid toxicity when all other explainations are ruled out. Bone marrow aspiration, liver biopsy and temporary cessation or decrease in lipid infusion should be discussed. Lipid infusion should be stopped until normalization of bilirubin level and platelet counts. Essential fatty acid deficiency has to be carefully checked.

Medium-chain based emulsions (50% MCT) might help to minimize the LCT load and theoretically their hepatic deposition [32-34] and long-term toxicity, but neither the current study nor other clinical data have demonstrated any effects of MCT- based emulsions to prevent or to reverse long term PN-associated cholestasis or hematologic complications [35-38]. Emulsions based on olive oil might also reduce the risk of lipid toxicity due to peroxydation, by decreasing the amount of linoleic acid [39,40].

Finally there is a relationship between lipid emulsions, macrophage activation syndrome, thrombocytopenia and cholestasis in a pediatric population homogeneous for liver disease risk factors. As lipids are indispensable in PN-dependent children, we recommend prevention through limitation of lipid supply and infusion rhythm to the patient's theoretical oxidizing capacity. The preventive role of long term use of emulsions countaining low proportion of polyunsaturated fatty acids needs to be explored in further studies.

References

  1. Macfie J, Smith RC, Hill GL. Glucose or fat as a nonprotein energy source? A controlled clinical trial in gastroenterological patients requiring intravenous nutrition.Gastroenterology 1981; 80: 103-7
  2. Nose O, Tripton JR, Ament ME. Effect of the energy source on changes in energy expenditure, respiratory quotient, and nitrogen balance during total parenteral nutrition in children. Pediatr Res 1987; 21: 538-41
  3. Pineault M, Chessex P, Bisaillon S, Brisson G. Total parenteral nutrition in the newborn: impact of the quality of infused energy on nitrogen metabolism. Am J Clin Nutr 1988; 47: 298-304
  4. Bresson JL, Bader B, Rocchiccioli F, et al. Protein-metabolism kinetics and energy-substrate utilization in infants fed parenteral solutions with different glucose-fat ratios. Am J Clin Nutr 1991; 54: 370-6
  5. Defronzo RA, Jacot E, Jequier E, et al. The effect of insulin on the disposal of intravenous glucose. Diabetes 1981; 30: 1000-7
  6. Stein PT, Mullen JL. Hepatic fat accumulation in man with excess parenteral glucose. Nutr Res 1985; 5: 1347-51
  7. Bresson JL, Narcy P, Putet G et al. Energy substrate utilization in infants receiving total parenteral nutrition with different glucose to fat ratios. Pediatr Res 1989; 25: 645
  8. Salas J, Girardet JP, De Potter S, Goulet 0, Ricour C. Glucose versus glucose-fat mixture in the course of total parenteral nutrition: effects on substrate utilization and energy metabolism in malnourished children. Clin Nutr 1991; 10: 272-8
  9. Pierro A, Carnielli V, Filler RM, et al. Metabolism of intravenous fat emulsion in the surgical newborn. J Pediatr Surg 1989; 24: 95
  10. Olivecrona G, Olivecrona T. Clearance of artificial triacylglycerol particles. Curr Opin Clin Nutr Metab Care 1998; 1: 143-151
  11. Carpentier YA. Intravascular metabolism of fat emulsions: the Arvid Wretling lecture ESPEN 1988. Clin Nutr 1989; 8: 115-25
  12. Carpentier YA, Simoens C, Siderova V et al. Recent developments in lipid emulsions : relevance to intensive care. Nutrition 1997; 13: 735-785
  13. Palmblad J. Intravenous lipid emulsions and Lost defense a critical review. Clin Nutr 1991; 10: 303-8
  14. Goulet O, Girot R, Maier-Redelsperger M, et al. Hematologic disorders followong prolonged use of intravenous fat in children. J Parent Enteral Nutr 1986; 10: 284-288
  15. Colomb V, Jobert-Giraud A, Lacaille F, Goulet O, Fournet JC, Ricour C. Role of Lipid Emulsions in Cholestasis Associated with Long-Term Parenteral Nutrition in Children. J Parent Enteral Nutr 2000; 24: 345-50

A complete list of references is available from the author.