Total Parenteral Nutrition (TPN)

Complications of parenteral nutrition

8.1 FLUID OVERLOAD 

 Infusion of large volumes of fluid as TPN obviously carries with it the danger of fluid overload and cardiac failure. This is particularly true in patients with pre-existing ischaemic or valvular heart disease and in patients with renal impairment, who are unable to excrete a water load. Under these circumstances one has to be particularly careful, particularly in the elderly. In borderline cases one might be able to manage with concurrent administration of a loop diuretic such as frusemide. If this is not adequate with florid pulmonary oedema, or the patient is oliguric, then the fluid will need to be removed either by haemodialysis or by haemofiltration. 

 8.2 HYPOVOLAEMIA 

 TPN often presents a large osmolar load to the circulation, which in the presence of normal renal function can lead to an osmotic diuresis with consequent volume and sodium depletion. This is more prone to happen if the total volume is infused over a relatively short period and can be mitigated if the duration of infusion is prolonged adequately. If the latter measure is not successful then one has to consider reducing the osmolar load of the fluid. Fortunately, this problem tends to be self limited. Sub-clinical hypovolaemia may only be detectable by measurement of postural fall in blood pressure, which makes postural measurements an important part of clinical assessment in these patients.

8.3 SODIUM IMBALANCE 

 Hyponatraemia may occur as a result of an osmotic diuresis produced by TPN, particularly when the feed contains inadequate amounts of sodium. This is even more likely to occur when diuretics are administered in an attempt to tackle hypoproteinaemic oedema. Routine monitoring of electrolytes is therefore important and the problem can be simply solved by addition of extra sodium chloride to the feed. Offending drugs must be discontinued. Persistent and unexplained low serum sodium levels must alert one to the possibility of pseudohyponatraemia, caused for example, by hyperlipidaemia. Hypematraemia has been encountered on occasion and is usually related to a hyperosmolar dehydration state secondary to the infusion of large volumes of highly concentrated dextrose, analogous to the situation that arises sometimes in elderly diabetics. The condition can be reversed by administration of isotonic or hypotonic dextrose. 

8.4 POTASSIUM IMBALANCE 

 Hypokalaemia can occur during TPN, on the same basis as hyponatraemia; osmotic diuresis, inadequate supplementation and diuretic excess. It is corrected in the same way by addition of potassium supplements. Hyperkalaemia is rarely a problem as long as renal function is normal. In patients with acute or chronic renal failure, this becomes a major problem, frequently encountered as a number of currently available parenteral solutions have variable quantities of potassium incorporated at the stage of manufacture. In patients with renal failure, therefore, it is important to check the potassium status before starting nutrition and, if necessary, use potassium free solutions. If there is a modest degree of renal function with a urine output of more than 1200 ml day-1, then hyperkalaemia may be managed by administration of loop diuretics such as frusemide. If the serum potassium is greater than 6.5 mmol 1-1 and/ or there is evidence of oliguria, then other measures will have to be tried (glucose-insulin infusion, sodium bicarbonate, calcium, beta-2 agonists, oral ion exchange resins and dialysis).

8.5 GLUCOSE DISTURBANCES 

 It is not unusual to find hyperglycaemia during TPN, and sometimes this can persist even after discontinuation, resulting in a true diabetic state. The causal factors include a rate of infusion higher than the body's ability to metabolize glucose, impairment of utilization by stress (trauma, infection), liver or pancreatic disease and drugs (cephalosporins, diazoxide, steroids). Hyperglycaemia can be ignored in the short term as it is likely to correct itself, but if it does persist after a week of TPN and if the blood concentration is persistently greater than 12 mmoll- 1 then treatment with insulin must be considered. There is still controversy regarding the question of whether or not to use insulin routinely during the acute phase of an illness when patients are commenced on TPN. Under the conditions described above it is reasonable to use low dose intravenous insulin infusion (1-6 units h- 1) with a target blood glucose of 6-12 mmol 1-1• It is worth emphasizing the dangers of hypoglycaemia. Rebound hypoglycaemia can occur following the cessation of feed, due to an insulin over-shoot. Fortunately this is rare and can be corrected by administration of dextrose. If the problem becomes recurrent then the feeding schedule must be altered so as to taper off the feed over a few hours, rather than sudden cessation.

8.6 ACID-BASE DISTURBANCES 

 Metabolic acidosis is unusual with the currently used TPN solutions, but has been reported in the past, related to administration of fructose, sorbitol or ethanol used as energy substrate, leading to accumulation of lactic acid. The use of dextrose and fat emulsion as energy substrate overcomes this problem. If it does occur for any reason, it can be corrected by administration of sodium bicarbonate. 

8.7 HYPOPHOSPHATAEMIA 

 This may occur due to inadequate phosphate supplementation in the feed, particularly when hypertonic glucose administration drives phosphate into the cells. Addition of adequate amounts of phosphate to the feed will correct the situation. 

 8.8 HYPERURICAEMIA 

 This was encountered when sorbitol, fructose or xylitol were used as energy substrates and is rarely seen today. 

 8.9 HYPERLIPIDAEMIA 

 Hypertriglycerideaemia (Type V hyperlipoproteinaemia) is a common problem associated with TPN, particularly in the long term. Hypercholesterolaemia is less predictable. The long term implications of these abnormalities with regard to increased cardiovascular risk is at present unknown. With increasing experience in the field definitive evidence might be forthcoming, with appropriate therapeutic implications. For patients on long term TPN we advocate regular screening of the lipid status and in the presence of persistent abnormalities treat them in the logical way one would treat any other patient: by reduction in the lipid component of the TPN to a minimum (500 m1 of Intralipid 10% twice a week) and control of obesity by reduction in the total caloric intake. It is not uncommon to see HPN patients who are over-nourished! We have not found it necessary so far to resort to lipid lowering agents, whose absorption may, in any case, be severely impaired by the lack of adequate bowel. 

 8.10 TRACE ELEMENT DEFICIENCIES 

 Medical science is rapidly learning more about the physiological role of trace elements, mainly due to experience with long term parenteral nutrition. Deficiency states (or toxicity) can only be established by sophisticated blood tests. Many of the currently available solutions incorporate variable amounts of some of these elements. Many elements occur as contaminants during manufacture and may even be present in excessive amounts - we have seen one example where chromium levels were very high. Additional supplements of specific trace elements are now commercially available, examples being zinc and selenium. With identification of specific roles and deficiency states, many more of these trace elements are likely to become commercially available. Clinical aspects have been dealt with in an earlier chapter. 

8.11 ESSENTIAL FATIY ACID DEFICIENCY 

 Linoleic and linolenic acids are both thought to be essential to man and need to be provided in adequate amounts in the diet; the latter is the precursor of arachidonic acid, which gives rise to prostaglandins. Though considerable amounts may be present in adipose tissue a state of deficiency (EFAD) can be precipitated by impaired mobilization secondary to hyperinsulinaemia produced by hypertonic glucose infusions. Arachidonic acid is a 20-carbon fatty acid with four double bonds and when the essential precursors are deficient, the use of oleic acid as substrate gives rise to the abnormal appearance of an alternative 20-carbon eicosatrienoic acid with three double bonds. EF AD was identified in man during the early days of TPN when fat emulsions were avoided (due to the adverse reactions to emulsions derived from com and castor oils). Today we have an eminently safe preparation which will ensure the prevention of this deficiency. It can still occur if the source of energy in the feed is solely carbohydrate in nature. Clinical features of EFAD include dry scaly skin, rash, loss of hair, platelet abnormalities and a neurological syndrome resembling sensorimotor neuropathy. The diagnosis is supported by the finding of an abnormally high trienoic to tetraenoic acid ratio in the plasma, which should be less than 40%. The abnormality is simply corrected by the inclusion of a fat emulsion which contains both substances, in the nutritional regimen. We recommend providing enough EFA to account for 5-10% of the total calories. In practice this can be achieved by the administration of as little as 2 1 of a 10% fat emulsion such as Intralipid, per week; we recommend at least twice that amount for routine maintenance nutrition (in the absence of significant hyperlipidaemia).   

8.12 THE CHINESE RESTAURANT SYNDROME 

 This is a syndrome of sudden and self-limited circulatory dysfunction manifested by pallor, sweating, tachycardia, tremors and hypotension. It is thought to be due to the rapid infusion of glutamic acid present in nitrogen based fluids. No specific treatment is indicated.  

8.13 ANAEMIA 

 In the earlier years of parenteral nutrition, when solutions based on linseed oil and cottonseed oil were used, the occurrence of a chronic low grade haemolytic anaemia was described. Nowadays the fat emulsions generally used are based on soya bean and this complication does not arise. Macrocytic anaemia has also been described in alcohol based feeds in the past, due to impaired folate metabolism. Again this does not apply as ethanol is no longer used as a source of energy - at least not parenterally! 

8.14 HYPERCALCAEMIA 

 The exact pathogenesis of this entity during TPN is not clearly understood. There is no demonstratable hyperparathyroidism or vitamin D toxicity. Fortunately it is self limited and has not been shown to produce any adverse effects. 

8.15 TPN OSTEODYSTROPHY 

 This curious syndrome has been predominantly documented in North America, occurring in patients on long term TPN. The symptoms include bone pains and pathological fractures. There are physical and radiological signs of osteomalacia. The exact nature of the disease is poorly understood. There are conflicting views on the role of vitamin D in this problem. There is some evidence to suggest that there may be resistance requiring large doses of the vitamin. On the other hand the syndrome is said to be reversed (paradoxically) by exclusion of vitamin D from the feeds and this has been attributed to vitamin D toxicity or excessive end-organ sensitivity. It is significant, however, that a similar picture may arise in patients with chronic renal failure on long term haemodialysis whose dialysis bone disease' is now attributed to aluminium intoxication.

In some cases of TPN osteodystrophy aluminium deposition

along the osteoid seam has been demonstrated and that

may indeed be the explanation for this disease. Further

research is required in this area and at present there is no

known treatment.

8.16 CHOLELITHIASIS

This is a recognized complication of long term TPN. The exact

mode of pathogenesis is unclear, but may relate to impaired

flow of bile or to altered bile salt/ acid metabolism. This can

lead to the usual complications such as cholecystitis and

obstructive jaundice.

8.17 INTRAHEPATIC CHOLESTASIS

This sinister complication is often a harbinger of death. It

usually occurs in patients on long term treatment. Often,

during the early weeks of TPN, a rise in the liver enzymes

(aspartate transaminase and alkaline phosphatase) is noted

and this seems to be a non-specific and harmless phenomenon.

Fatty liver may be seen on histology and in itself may

not be a clinical problem.

However, the appearance of jaundice with a persistently

raised serum bilirubin and features of obstructive jaundice, in

the absence of demonstrable extrahepatic obstruction by gall

stones, is an ominous sign. The condition often progresses to

liver failure and death. Histology of the liver reveals a picture

of intrahepatic cholestasis. An aetiological agent has not been

clearly delineated, though there have been suggestions that it

may relate to conditioned deficiency of taurine or choline.

There is the real possibility that it may be the result of inadvertent

toxicity caused by excessive infusion of some obscure trace

element.

There is no known treatment, apart from cessation of TPN

and reversion to oral feeding. Exclusion of specific components

such as lipids, or change to alternative preparations have

proven futile, though worth a try. Steroids have not been of

any benefit in our experience. The potential role of trace

elements (deficiency or toxicity) in this condition is worthy of

serious consideration.

Miscellaneous deficiencies of metabolites have been suggested

or identified during TPN - tyrosine, choline, taurine, cysteine

and carnitine. The exact significance has not been established

firmly and requires further investigation