Total Parenteral Nutrition

Aims

• To provide the nutritional requirements for optimal growth and maturation of the infant. It is a substitute for enteral feeding in circumstances where the establishment of full enteral feeds will be delayed. However, the preferred form of nutrition for the neonate remains breast milk ¹ .

Why is it necessary ?

• Many low birth weight babies cannot be fed enterally to their full requirements in the first weeks of post natal life. Unless they are provided with the appropriate amounts of carbohydrate, fat and protein they will inevitably be catabolic resulting in a negative nitrogen balance. This early deficit may be of significance regarding future growth, resistance to infection and neuro-developmental outcome.

• Pre-natal enteral nutrition (of amniotic fluid) maintains gut development in utero, and work has shown that luminal nutrition is essential for the maintenance of gut structure and function ² . These concerns have prompted the evaluation of non-nutritive feeding (NNF) or minimal enteric feeding (MEF) as an intervention to aid gut maturation. A trial of TPN versus TPN with minimal enteric feeds resulted in better growth (as weight gain), shorter hospital stay and less osteopaenia of prematurity in the group receiving MEF in addition to TPN ³ .

• TPN has reached its current level of use, and perceived benefit, despite a lack of randomised controlled trials (RCT’s) comparing TPN to other forms of nutrition in matched populations. Studies are few that prove TPN decreases either morbidity or mortality and are now unlikely to be undertaken as it would be deemed unethical to withhold TPN from a control group.

Indications

• Premature infants <30 weeks gestation and/or <1000g.
• >30 weeks gestation but unlikely to achieve full enteral feeds by day 5.
• Severe inter-uterine growth restriction.
• Necrotising enterocolitis (NEC).
• Gastro-intestinal tract anomalies.

Components of TPN

• Fluid. Fluid is an essential component of parenteral nutrition. Volumes are increased over the first 7 days in line with the fluids and electrolytes protocol with the aim of delivering 150 mls/kg/day by day 7.
  

• Calories. Babies need 100-120 kcal/kg/day to grow, some babies will need more that this. In TPN this is provided as glucose (0.4kcal/ml @ 10%) and lipid (1.1kcal/ml @ 10% and 2.0kcal/ml @ 20%). Protein, although a potential energy substrate, should only be utilised for tissue growth with the former two energy sources provided in sufficient quantity to avoid protein catabolism. In preterm infants, unless the glucose and lipid amounts are pushed beyond the usual limits of tolerance, it is difficult to deliver sufficient calories for good growth with TPN. Total daily calorie delivery will usually be between 85 and 95 kcal/kg/day. This will prevent catabolism and provide calories for some rather than optimal growth.

• Carbohydrate. Glucose is the main source, usually provided as 10% Dextrose. Premature infants are often relatively intolerant with glycosuria (but not always an osmotic diuresis). If this problem arises the use of an insulin infusion obviates the need to decrease the concentration of dextrose prescribed, and will also optimise the utilisation of calories by the infant ⁴ . Minimal enteric feeds, with their trophic action on gut hormones also allows a higher concentration of glucose to be tolerated .

• Protein. This is delivered as a synthetic crystalline amino acid solution. This is currently delivered as Vamin N but we will soon be moving to Primene. McIntosh et al showed the latter solution produced blood amino acid levels closer to cord blood reference levels than Vamin. ⁵   There are 9 essential amino acids with cysteine, tyrosine, taurine and arginine in addition as the semi-essential group of amino acids. In the absence of exogenous protein, a preterm infant will catabolise 1g/kg/day of their own body protein to meet their metabolic needs. If 1-2g/kg/day of protein is provided, along with at least 70kcal/kg/day of non-protein calories, catabolism will be prevented. Therefore the prompt introduction of amino-acids via TPN achieves an early positive nitrogen balance for the infant, the ideal being an accretion of 200 to 300mg/kg/day ^6,7 . We would aim to deliver 3g/kg/day of protein although there is observational evidence that very preterm babies will tolerate upto 4g/kg/day without adverse biochemical or clinical consequences. ⁸ Recent work, so far only published in abstract form ⁹   suggests it is as safe to start babies immediately on 3 g/kg/day as building up the protein load slowly over 3 days). Adverse effects of excess protein include a rise in urea and ammonia and high levels of potentially toxic amino acids such as phenylanaline. TPN can also casue metabolic acidosis which can be modified by the addition of a buffer, acetate. In one RCT the partial replacement of chloride by acetate in the amino acid solution resulted in an improved pH, a reduction in both bicarbonate and colloid use, with no adverse effect on ventilation requirements compared to the group receiving standard TPN ¹⁰ .

• Intralipid. An oil-in-water emulsion derived from egg phospholipid, soyabean and glycerol. An excellent source of energy, and therefore nitrogen sparing. Lipid is started immediately at 1 g/kg/day and increased to 3g/kg/day in daily increments of 1g/kg/day. The Soybean provides the essential fatty acids, linoleic and alpha-linolenic, although the latter is only present in a small quantity (8%). Triglyceride levels should be kept at < 150mg/L .¹¹ Both 10% and 20% formulations are available, the 20% formulation is cleared more efficiently due to a more favourable phospholipid : triglyceride ratio ¹² . TPN lipid is thought to be cleared from the blood stream in a similar manner to endogenous chylomicrons and is dependent upon lipoprotein lipase activity ¹³ . This can be augmented by the addition of heparin to the lipid infusion, although the liberated free fatty acids may not be cleared rapidly ¹⁴ . It is well tolerated over a 24hr infusion period ¹⁵ , although whether a lipid free interval is required for cyclical regeneration of the lipases is not clear.  Lipid exposed to light forms potentially toxic lipd hydroperoxides, so lipid syringes and tubing should be protected from light by wrapping in foil. ¹⁶

• Minerals. Sodium, potassium, chloride, calcium, magnesium and phosphorus levels need to be closely monitored and prescribed accordingly. Calcium and phosphorus provision can be difficult in the premature infant due to the solubility problems in small volumes. The past practice of a 20hrs on, 4 hours off pattern for lipid infusion was necessary for the measurement of plasma electrolytes. Analysers could not cope with lipaemic serum but this is no longer a problem with modern equipment. Carnitine is essential for fatty acid oxidation (it facilitates transfer across the mitochondrial membrane). Premature infants have low stores of carnitine and supplementation may improve fat utilisation. At present it is not routinely added to parenteral solutions ¹⁷

• Trace Elements. Zinc, copper, manganese, selenium, fluorine and iodine are provided in a number of commercial TPN preparations. If not present they should be added for TPN of greater than 7 days duration. Levels are usually only checked where TPN has been required for greater than 3 weeks. Other trace elements such as molydbenum and chromium may also need to be assayed. It may also be possible to add iron to TPN in the near future.

• Vitamins. The daily requirements for both water and fat soluble vitamins can be provided in TPN. Degradation by ambient light necessitates the covering of the amino acid and lipid solutions either with aluminium foil or the use of an opaque tubing. ¹⁸   Adding multivitamins preparations to the intralipid seems to reduce light induced formation of lipid hydroperoxides

Prescription

Premature infants tolerate TPN from day 1 of post-natal life. By commencing TPN at this time it avoids the problem of endogenous protein catabolism. Recent work suggests that early TPN can improve nitrogen retention and blood glucose homeostasis, without an increase in complications such as jaundice, acidosis or lipid intolerance.¹⁹   Although growth (and energy intake) was increased in the early TPN group there were no significant differences in survival, broncho pulmonary dysplasia (BPD), NEC, cholestasis, osteopaenia, sepsis or duration of hospital stay.

Parenteral nutrition can be delivered using standardised or individualised bags. There is observational evidence that most TPN prescription can be adequately done using standardised bags. ²⁰ Because standardised bags are both easier and cheaper, this is the preferred method on this unit. Some babies, particularly is they are very unstable, may need individualised bags and there is a computer program within the database to aid with these individualised prescriptions. Individualised prescription should only be done after consulation with the consultant on service.

Standardised Bag Prescribing.

• The amino acid bags will be left connected for 48 hrs.
  
• The nutrient intake in the standardised bags is optimised at 150 mls/kg/day, so as fluid volumes are graded up during the first week the nutrient intake will be lower than this. This should not be a problem unless fluid restriction persists beyond the first week of life.
• Sodium is the commonest mineral that needs manipulation in individualised prescribing of TPN. If the sodium falls on 10% Preterm TPN Normal Sodium then 10% Preterm TPN High Sodium can be used. This will deliver twice as much sodium, 9 mmol/kg/day at 150 mls/kg/day.
• Adding other Trace Elements should be considered if TPN remains the dominant source of nutrition after 2 weeks.
  

Individualised Bag Prescribing.

• Prolonged fluid restriction.
• Severe or prolonged sodium losses, use 10% Preterm High Sodium first.
• Glucose intolerance resistant to insulin. 
• Evidence of high blood levels of amino-acids.
• Any others?

Administration

TPN should be delivered where possible through central lines. Peripheral lines are only suitable for TPN of less than 3 days duration with restrictions on the dextrose concentration prescribed. Usually a percutaneous central line is placed with the position of the tip of the catheter confirmed on x-ray prior to use. The amino acid solution is attached to a burette with a suitable bacterial filter in line before a "Y" connector to which the lipid infusion is attached (?lipid filters). A strict aseptic technique in preparation and administration of the TPN is essential. Ideally, breakage of the central line through which the TPN is infused should be avoided, though compatible drugs can be administered if necessary. 

Cautions

• Hyperkalaemia. Addition of potassium is rarely required in first three days of life unless the serum potassium is < 4.00mmol/l. Also use caution when prescribing in renal impairment. A minimal amount is inevitable in TPN because of the type of amino acid formulation used.

• Hypocalcaemia. May result from inadvertent use of excess phosphate. Corrects with reduction of phosphate.

• NEVER add bicarbonate, it will precipitate calcium carbonate out.

• NEVER add extra calcium to the burette, it will precipitate out the phosphate.

• Toxicity due to accumulation of certain amino acids should be considered in an infant becoming unwell and acidotic on TPN. A urinary amino acid screen is required.

• Fatty acids. Due to fatty acids being precursors of prostaglandin synthesis potential adverse effects on pro/anti-coagulation homeostasis and pulmonary vasculature tone are theoretically possible.

Complications

1. Delivery

The line delivering the TPN may be compromised by;

• Sepsis, minimised by maintaining strict sterility of the line during and after insertion. This problem was addressed in a study which added low dose vancomycin to the TPN ²² . There was a reduction in the number of coagulase-negative staphylococcal bacteraemias in the vancomycin group as was the length of hospital stay. There were no reported vancomycin resistant strains isolated during the study but concerns regarding this possibility prevented its incorporation into clinical practice.
• malposition, x-ray mandatory before infusion commences.
• thrombophlebitis, with peripheral lines, requiring close observation of infusion sites.
• extravasation into the soft tissue, with resulting tissue necrosis.

2. Metabolic complications

• Hyperglycaemia
• Hyperlipidaemia
• Cholestasis

Hyperglycaemia can be controlled effectively with an insulin infusion as opposed to reducing the glucose concentration. Minimal enteric feeds also have a glucose lowering effect. Unutilised carbohydrate can be converted into endogenous lipid resulting in a fatty liver. This too can be ameliorated by both enteral feeds and the use of insulin. Cholestasis is well recognised either due to hepato-toxicity of the infusate or the lack of hepatic stimulation in the absence of enteral feeding. Typically it is only present when TPN has been required for several weeks.

3. Potential adverse effects of lipids

Lipid administration has been the subject of much research regarding in particular the questions of its safety and possible adverse effects on other organs.

1. Pulmonary function.

   It has been suggested that exogenous lipid interferes with respiratory function. This was based on observations of a 7-fold increase in chronic lung disease over a period when use of intralipid became widespread ²³ . However, a prospective study failed to demonstrate a significant association between intravenous lipid and chronic lung disease ²⁴ . Suggested mechanisms include impaired gas exchange from pulmonary intravascular accumulation or impaired lymph drainage resulting in oedema. Parenteral lipid may also shift the prostaglandin synthesis pathway in favour of pulmonary vasoconstrictors (eg thromboxanes). There are reports of increased pulmonary vascular resistance of a dose and time dependent nature which suggest lipid may aggravate pulmonary hypertension in susceptible individuals ²⁵ . It is thought that only the higher rates of lipid infusion (>2g/kg/day) produce these changes, particularly with the intermittent infusion regime.

2. Kernicterus

   Lipid itself does not displace bilirubin, but the liberated free fatty acids displace bilirubin from albumin. Some Units use the free fatty acid to albumin ratio as a guide with a value <6 taken as safe. In the presence of jaundice requiring phototherapy the higher concentrations of lipid (>2g/kg/day) should be avoided.

3. Thrombocytopaenia

   There has been a question for some time as to whether TPN lipid is damaging to platelets. Only one report has demonstrated thrombocytopaenia with TPN and these were children on long term home TPN for months to years. ²⁶ No causal relationship has been demonstrated in neonates. Indeed the absence of lipid in TPN can cause thrombocytopaenia through essential fatty acid deficiency.

4. Sepsis

   There are conflicting reports that lipid interferes with immune function, previously explained as the result of phagocytes accumulating fat which impaired their ability to function normally. These concerns regarding macrophage function in the presence of lipid were mainly in vitro studies not substantiated by later clinical trials. A study of ill neonates receiving 1g/kg intralipid demonstrated no change in polymophonuclear leukocyte count, chemokinesis, chemotaxis, aggregation, platelet count and aggregation pre and post infusion of lipid ^{27, 28} .

5. Free radical formation

   Lipid peroxidation of the polyunsaturated fatty acids occurs if they are exposed to light with an accelerated reaction during phototherapy. The adverse effect of these products of peroxidation are not fully elucidated as yet ¹⁸ . Covering with silver foil, using opaque tubing, or the addition of ascorbate prevents oxidation.

Monitoring

TPN administration requires careful clinical and laboratory monitoring. Adequate growth is best determined by linear growth as weight gain can reflect an increase in total body water rather than tissue accretion. A technique such as knemometry may be suitable for this purpose ²⁹ . In addition to routine observations the following are required for short term TPN use.

1. Laboratory
   • Full blood count, plasma sodium and potassium, creatinine.
     Required daily for 1 week then 3 times a week

   • Plasma calcium, magnesium, and phosphate.
     Twice a week until stable then weekly

   • Lipid levels.
     Twice a week first week then weekly unless complication arises (sepsis etc)

   • Long term TPN (> 2 weeks duration) requires, in addition, liver function tests and trace element assays.

2. Clinical
   • Blood sugar, 4-6hrly first 3 days, twice a day once stable.

Filtration

Conclusion

Parenteral nutrition fulfils a desire to supply nutrients to infants who cannot initially be fed enterally . This is because we believe starvation to be detrimental. It certainly permits faster weight gain but beyond that, although it appears to be well tolerated, trials demonstrating long term benefits of parenteral nutrition are not evident to date. Continued improvements in TPN formulation, and its constituent elements, will doubtless be made, perhaps at a greater prescription cost. It is imperative therefore that further trials are instigated in order to support the policy of parenteral feeding. To balance this "high tech" option it is noteworthy to mention that other centres propose a "low tech" approach using donor human milk banks to supply breast milk to their premature infants, greatly reducing the need for TPN and central venous access.

Key Points

Parenteral feeding results in earlier and faster weight gain (as tissue accretion) vs delayed enteral feeds alone (no change in mortality rates, no late outcomes reported)	11
Intensive parenteral regimes, without increasing adverse clinical outcomes, result in better growth (but no difference in morbidity or hospital stay)	6
Minimal enteric feeds with TPN leads to earlier tolerance of enteral feeds, and faster weight gain than TPN alone	3
Insulin infusion for TPN related hyperglycaemia is safe and results in increased glucose utilisation and hence faster weight gain	4
Individualised TPN bags are only neccesary in a minority of newborn prescriptions.	20

References

1. The Australian College of Pediatricians. J Paed Child Health 1998; 34: 412-13

2. Heird. Gastro-intestinal function and neonatal nutrition Ross Laboratories1977; 16

3. Dunn et al Beneficial effects of early hypocaloric enteral feeding on neonatal gastro-intestinal function: preliminary report of a randomized trial.J Pediatr 1988; 112:622-29

4. Collins et al. A controlled trial of insulin infusion and parenteral nutrition in extremely low birth-weight infants with glucose intolerance. J Pediatr 1991; 118: 921-27

5. McIntosh M, Mitchell V. A clinical trial of two parenteral nutrition solutions in neonates. Archives of Disease in Childhood 1990;65:692-99

6. Wilson et al Randomised control trial of an aggressive nutritional regime in sick very low birth-weight babies. Arch Dis Child 1997; 77F: 4-11

7. Rivera et al Plasma amino acid profiles during the first three days of life in infants with respiratory distress syndrome: effect of parenteral amino acid supplementation.J Pediatr 1989; 115: 465-68

8. Porcelli PJ, Sisk PM. Increased parenteral amino acid administration to extremely low birth weight infants during early postnatal life. Journal of Pediatric Gastroenterology and Nutrition 2002;34:174-9

   
   

9. Paisley JE, Thureen PJ, Baron KA, Hay WW. Safety and efficacy of low vs high parenteral amino acid intakes in extremely low birth weight neonates immediately after birth. Pediatr Res 2000; 47:293A, Abstr no: 1732

10. Peters et al Randomised control trial of acetate in preterm neonates receiving parenteral nutrition. Arch Dis Child 1997; 77F: 12-15 

11. Yu et al Total parenteral nutrition in very low birth-weight infants: a controlled trial. Arch Dis Child 1979; 54: 653-61

12. Haumont. Effect of liposomal content of lipid emulsions on plasma lipid concentrations in low birth weight infants receiving parenteral nutrition. J Pediatr 1992; 122: 759-6

13. Coran et al The role of fat in intravenous feeding of the newborn. J Paed Surgery 1974; 9: 725-732

14. Spear et al Effect of heparin dose and infusion rate on lipid clearance and bilirubin binding in premature infants receiving intravenous fat emulsions J Pediatr 1988; 112: 94-98

15. Kao et al Triglycerides, free fatty acids / albumin molar ratio, and cholesterol levels in serum of neonates receiving long-term lipid infusions: Controlled trial of continuous and intermittent regimes. J Pediatr 1984 (104) pp429 J Pediatr 1984; 104: 429

16. Silvers KM, Sluis KB, Darlow BA, McGill F, Stocker R, Winterbourn CC. Limiting light induced lipid peroxidation and vitamin loss in infant parenteral nutrition by adding multivitamin preparations to Intralipid. Acta Paediatrica 2001;90:242-249
    

17. Cairns and Wilson. Carnitine supplementation of parenterally fed infants. Cochrane Library 1997; review: (protocol). 

18. Neuzil. Oxidation of parenteral lipid emulsion by ambient and phototherapy lights: potential toxicity of routine parenteral feeding. J Pediatr 1995; 126:  785-90

19. Murdock et al Low birth-weight infants and total parenteral nutrition immediately after birth 2. Randomised study of biochemial tolerance of intravenous glucose, amino acids and lipid.feeding. Arch Dis Child 1995; 73F: 4-16

20. Beecroft C, Martin H, Puntis JW. How often do parenteral nutrition prescriptions for the newborn need to be individualised. Clinical Nutrition 1999;18:83-5

21. Matlow AG, Kitai I, Kirpalani H et al. A randomised trial of 72 versus 24 hour intravenous tubing set changes in newborns receiving lipid therapy . Infection Control and Hospital Epidemiology 1999;20:487-93
    

22. Baier et al Selective use of vancomycin to prevent coagulase-negative staphylococcal nosocomial bacteremia in high-risk very low birth-weight infants. Paed Inf Dis J 1998; 17: 179-183

23. Cook et al Factors associated with chronic lung disease in preterm infants. Arch Dis Child 1991; 66: 776-79

24. Aldwaidh et al. Randomised trial of effect of delayed intravenous lipid administration on chronic lung disease in preterm infants. J Pediatr 1996; 22: 303-06

25. Prasertsom. Pulmonary vascular resistance during lipid infusion in neonates. Arch Dis Child 1996; 74F: 95-98

26. Goulet et al. Haematological disorders following prolonged use of intravenous fat emulsions in children. J Parenter Enteral Nutrition 1986; 10: 284

27. English et al. Intravenous lipid emulsions and human neutrophil function. Pediatrics 1981; 99: 913-6

28. Herson et al. Effects of intravenous fat infusion on neonatal neutrophil and platelet function. J Parenter Enteral Nutrition 1989; 13: 620-22

29. Gibson et al. Knemometry and the assessment of growth in premature babies. Arch Dis Child 1993; 69: 489-504

30. Puntis JWL et al. Hazards of parenteral treatment: Do particles count. Arch Dis Child 1992;67:1475-77

31. Bethune K et al. Use of filters during preparation and administration of parenteral nutrition: Position paper and guidlines prepared by a British pharmaceutical nutrition group working party. Nutrition 2001;17:403-8

32. Van Lingen RA et al. ELD96 particle filters in sick newborn infants result in significantly fewer infections at lower cost. J Clin Microbiol Infect 1997;3:122

33. Kunac DL et al. In line intravenous filtration in neonates: Help not hindrance . Australian Journal of Hospital Pharmacy 1999;29:321-7