This site is intended for health professionals only

Stability of parenteral nutrition mixtures

teaser

Mike C Allwood
BPharm PhD
FRPharmS GB
Emeritus Professor of Clinical Pharmaceutics
University of Derby
UK
E:[email protected]

To provide adequate intravenous nutritional support, mixtures containing upwards of 50 separate ingredients may be required. The opportunities for chemical and physical interactions between any of these chemical components are substantial. An understanding of the most significant causes of such reactions is necessary  to ensure that a safe and efficacious parenteral nutrition (PN) mixture can be compounded. It is the pharmacist’s role to assess any PN mixture for pharmaceutical interactions, to minimise chemical change and to decide on a maximum shelf-life.

Instability can be manifest in two general ways. Physical changes result in the formation of precipitates or the cracking of fat emulsions, while chemical instability results in a loss of biological activity.

Physical instability
The consequence of the formation of insoluble particles in parenteral mixtures by chemical interaction between different chemical ingredients is life-threatening and must be prevented.

Formation of divalent phosphate salts is the greatest risk, caused by poor solubility of certain divalent phosphate salts, particularly calcium phosphate, but also magnesium phosphate.(1) So any PN mixture containing inorganic phosphate salts and calcium or magnesium poses a potential risk of precipitation. But this is not inevitable. In practice, adding normal adult requirements for calcium and phosphate will not result in precipitation. The greater challenge concerns achieving the higher requirements of small children. The factors influencing precipitation are multifactorial and are summarised in Table 1.

[[HPE21_table1_52]]

Understanding the key issues affecting solubility of calcium phosphate allows us to ensure that risk is minimised and clinical requirements are met, by taking certain precautions. For example, using calcium gluconate rather than chloride reduces the likelihood of precipitation, but note that calcium gluconate may be heavily contaminated with aluminium, which is contraindicated in small children. Using the more acidic (monobasic) phosphate salt also reduces risk, because calcium dihydrogen phosphate is 60 times more soluble than the dibasic salt. But the safest method is to use an organic phosphorus source, such as sodium glycerophosphate, and this should always be the first choice for children, in combination with calcium chloride, the latter having the added advantage of being more concentrated.
In summary, the risk of calcium phosphate precipitating can be reduced by:

  • Using monobasic (acid) phosphate.
  • Adding cysteine to lower pH.
  • Using calcium gluconate, not chloride.
  • Adding phosphate before calcium salt.
  • Using an organic phosphate source.

Trace element microprecipitates can also form in certain PN mixtures. The important variable in this regard is the source of amino acid mixture used. Two examples are well recognised.(2) First, PN mixtures using Vamin‚ or Vaminolact‚ as the amino acid source contain small amounts of hydrogen sulphide as a contaminant. Consequently, the addition of copper salts, as found in all trace element additives, results in the slow formation of insoluble copper sulphide. This microprecipitate is usually invisible but can be isolated by filtration. It requires at least seven days before the precipitate can be detected, so restricting shelf-lives in Vamin-containing mixtures is essential.

Insoluble iron phosphate may be formed by reaction between iron salts in trace element additives and inorganic phosphates. This is a problem normally restricted to Synthamin-containing mixtures, and it is related to the presence of phosphate in the Synthamin infusion. It is slow-forming, requiring at least 14 days to be evident. It is avoided by using an organic phosphate source with electrolyte-free Synthamin‚ or by inclusion of ascorbic acid alone or as part of a multivitamin additive in the mixture, which has been shown to prevent this reaction.

Fat emulsion stability
Most PN mixtures are now of the all-in-one type, fat emulsion being included to permit one infusion a day. But fat emulsions are not necessarily stable after addition to a PN mixture. A number of factors affect fat emulsion stability, and these are summarised in Table 2.(1)

[[HPE21_table2_52]]

Maximum limits for divalent cations are particularly important, as are the proportion of fat by volume and the minimum and maximum concentration of glucose. These limits vary with the amino acid and fat emulsion source. Each manufacturer provides information describing such limits.

Chemical instability
Most ingredients are remarkably stable after PN compounding, the main exception being certain vitamins. The least-stable vitamins, together with reasons for poor stability, are summarised in Table 3.(2)

[[HPE21_table3_53]]

Degradation can occur in two situations: first, after compounding; and secondly, during infusion. The least-stable ingredient during storage of complete PN mixtures is ascorbic acid, which reacts with dissolved oxygen, a reaction catalysed by copper ions. This reaction is rapid and is often complete in less than 48 hours. The quantity of ascorbic acid degraded depends on the amount of dissolved oxygen in the infusion. This oxygen originates from a number of sources, including ingredients (eg, glucose infusion and microadditives, but not amino acid infusion or fat emulsion), aeration during filling and gas permeation through EVA bag walls (avoided by using multilayered bags). In a typical adult regimen, losses of ascorbic acid will amount to 50–75mg within 24 hours of compounding, but no further losses occur if in a multilayered bag. If EVA is used, all the ascorbic acid will be removed within 72–96 hours. So vitamins can be added to PN mixtures during compounding in multilayered bags and extended shelf-lives assigned, often at least 14 days. But a mixture in EVA must be restricted to 72 hours. Multichambered bags are oxygen-free until the outer wrap is removed, so a slightly longer shelf-life is possible after compounding, compared with EVA.

Photodegradation during infusion is also likely. Vitamin A is the most light-sensitive vitamin, degrading rapidly in daylight; most if not all vitamin A is lost in an hour or two, depending on daylight intensity and type of mixture. Fat emulsion provides some but by no means complete protection. Covering the bag is essential, but degradation still occurs during passage through the set. Covering the set would be ideal. Administration at night or only in artificial light conditions avoids photodegradation, as vitamin A and other light-sensitive vitamins are not degraded by non-UV-emitting light sources.

Conclusion
PN mixtures compounded by pharmacy, containing all the nutritional needs of patients, are remarkably stable, although some important physical and chemical changes can occur, most of which are avoidable. Extended shelf-lives can be assigned to complete PN mixtures, ensuring that the patient’s nutritional requirements are met while at the same time avoiding the dangers inherent in ward staff making final additions to PN mixtures before or during infusion.

References

  1. Allwood MC. Parenteral nutrition formulations. In: Paynes-James J, Grimble GK, Silk DBA, editors. Artificial nutrition support in clinical practice. 2nd ed. London: Greenwich Medical Media; 2001. p. 435-44.
  2. Allwood MC, Kearney MC. Compatibility and stability of additives in parenteral nutrition admixtures. Nutrition 1998;14:697-706.

Resources
British Pharmaceutical
Nutrition Group (BPNG)
W:www.bpng.co.uk
British Association of Parenteral and Enteral Nutrition (BAPEN)
E:[email protected]
European Society of Parenteral and Enteral Nutrition (ESPEN)
W:www.espen.com

Events
BPNG Summer Symposium
18–19 June 2005
Leicester, UK
W:www.bpng.co.uk
E:[email protected]
BAPEN Annual Meeting – Advanced Clinical Nutrition
16–17 November 2005
Telford, Sovereign Conference, Redditch, UK
ESPEN Congress
27–30 August 2005
Brussels, Belgium
W:www.espen.com






Be in the know
Subscribe to Hospital Pharmacy Europe newsletter and magazine

x