Send letter to journal:
Re: Solutions are less complicated than theory.

We read Dr Coulthard’s article on 0.18% saline/4% glucose with interest.

Alert 22 form the NPSA did not recommended 0.45% sodium chloride with 2.5% glucose by default. Firstly the NPSA did not designate a “default solution” per se. Secondly their advice gives a near exhaustive list of conditions in which ‘children should only ever be prescribed isotonic solutions” and then goes on to recommend 0.9% saline with or without glucose and Hartman’s solution. Therefore if one were to choose a default solution based on Alert 22, it would logically be one of these more concentrated solutions (1).

Even if applying Dr Coulthard’s rather complex formula were practicable, and we would question some of its basic assumptions, it is a mistake to portray as equivalent the risks of hypernatraemia - which is nearly always well tolerated even when severe - to those of hyponatraemia which has caused brain damage and death at levels within 15 mmol/L of the physiological range (2,3,4). Furthermore the strength of the NPSA’s guidance is that it encourages sound clinical and biochemical monitoring rather than procedures based on physiological nuance and intricacy.

We suggest that Dr Coulthard’s formula, owes any success it may enjoy to his initial use of 0.9% saline where many would just progress to routine maintenance with a more isoplasmic solution in the first place. Like many doctors, our experience is that electrolytes behave predictably when infused. We audited data for three years on perioperative children receiving 0.45% sodium chloride with 2.5% glucose (n= 57), 0.9% saline with 5% glucose (n=53) and Hartmann’s solution with 3% glucose (n=41). Looking at serum electrolytes 12 hourly, we found that 0.45% sodium chloride with 2.5% glucose had little impact on the serum sodium, such that roughly 25% of patients were hyponatraemic initially and throughout all four subsequent samples. Indeed Dr Coulthard’s assertion that 0.45% will increase the incidence of hypernatraemia is not borne out by any published clinical series or our own audited data. Our experience with 0.45% saline, was that its sodium content was insufficiently high to prevent hyponatraemia.

Extending Dr Coulthard’s logic surely one might expect to see a dramatic rise in serum sodium when 0.9% sodium chloride is given? Again looking at our own audited data, this did not occur. While we observed a rise in mean sodium from 137mmol/L (SE = 2.8) initially to 139 mmol/L (SE = 3.0) 48 hours later, no patients became hypernatraemic at any stage. We changed our institutional default to Hartmann’s solution with glucose 3% having observed high levels of hyperchloraemia (greater than 50% of patients) with 0.9% saline and 5% glucose, and this was broadly successful.

So while it may be refreshing to read a defence of a now little-used solution, there has been a disturbing tendency abroad to deny that there were problems with the widespread use of 0.18% sodium chloride with 4% glucose. There may well be a rationale for using it within certain constraints but the fact is that the track record of this substance in routine paediatric use has been nothing short of disastrous and it is unsafe for general use.

References

Send letter to journal:
Re: Fluid prescription should be logical, safe and simple.

We were interested to read Dr Coulthard’s opinions on maintenance intravenous fluids(1) in the context of NPSA patient safety alert 22.(2) As highlighted by Stewart et al in their response to Dr. Coulthard’s article(3) there is no default fluid designated by the NPSA however when we examined all of the NPSA recommendations, with regard to our patient population, we decided to opt for the isotonic fluid 0.9% Saline with 5% Dextrose as a single “default fluid”. We believe that this is logical, safe and simple.

Firstly it is a logical fluid. The majority of children in whom intravenous fluids are required on a general paediatric medical and surgical ward are likely to have excess ADH secretion, and therefore 0.9% Saline with 5% Dextrose is appropriate. In this circumstance we also decrease the fluid volume to two-thirds of the volume traditionally cited as “maintenance”, except in the dehydrated child.

Secondly it is safer than a hypotonic fluid. In children with excess ADH secretion, there is a tendency to develop hyponatraemia. This is not prevented by use of an isotonic fluid but it would take longer to develop than with infusion of a hypotonic fluid, giving opportunity for detection before serious sequelae occur. Rigorous monitoring of electrolytes is an essential part of fluid prescription, again as emphasised in the NPSA recommendation, and the algorithm described in Coulthard’s article is a useful one in the children in whom hyponatraemia develops.

Finally it is simple – further contributing to safety. Fluids may be prescribed by colleagues who are not familiar with paediatric patients; we have minimised fluid prescribing in children by non-paediatric doctors. They may be commenced at night or at exceptionally busy times, where human error is more likely. Reducing complexity is a major factor in changing systems to prevent patient harm.(4)

There are potential complications with this fluid, such as hypernatraemia, hyperchloraemia, metabolic acidosis, hypertension and phlebitis, and we are currently auditing this; however we believe the evidence is that the risk of these are less than the risk of hyponatraemia. We recognise the risks of adopting a “new standard of care” without a good evidence base and we support the call by Choong et al, following their systematic review of hypotonic vs isotonic fluids(5) , for clinical trials to compare different fluid regimens. In the meantime a national clinical alert network should be established to report problems with the administration of isotonic fluids – such a facility may have identified the problems with hypotonic solutions with which we are only dealing with now – over 50 years later.

In summary, the use of a single default isotonic fluid such as 0.9% Saline with 5% Dextrose is logical, safe and simple, when used together with fluid restriction and close monitoring as recommended by NPSA. Patient-tailored prescriptions can follow later if required.

References

(1)Coulthard MG. Will changing maintenance intravenous fluid from 0.18% to 0.45% saline do more harm than good? Arch Dis Child 2008 ;93(4):335-40.

(2) NPSA Safety Alert 22. Reducing the risk of harm when administering intravenous fluids to children. National Patient safety Agency. http://www.npsa.nhs.uk. 2007

(3)Stewart PC, Nutt C, McGrath K. Solutions are less complicated than theory. Arch Dis Child 2008, e-letter.

(4) Nolan, T. System changes to improve patient safety. BMJ 2000;320:771-773.

Send letter to journal:
Re: Isotonic maintenance fluids do not produce hypernatremia

Coulthard proposes that 0.18% saline should be the maintenance fluid of choice in children and that an intravenous fluid with a sodium composition of 0.45% saline or higher will lead to an increased number of cases of hypernatremia. His arguments are based primarily on two false premises: 1) that SIADH is extremely rare and 2) that renal concentrating defects are common. In 2003, we proposed the use of 0.9% saline as a maintenance parenteral fluid for the prevention of hospital-acquired hyponatremia, as hypotonic fluids were associated with multiple cases of death and neurological injury from hyponatremic encephalopathy in children.[1] Hospitalized children have multiple non-osmotic stimuli for AVP production that put them at risk for hyponatremia, such as nausea, vomiting, pain, stress, pulmonary disorders, CNS disorders, malignancies, narcotics, NSAIDS, dehydration, hypotension, sepsis, adrenal insufficiency and hypoalbuminemia. These children are all at risk for hospital-acquired hyponatremia and would benefit from 0.9% saline. Dr. Coulthard’s criteria for the diagnosis of SIADH, a urine osmolality > 600 mOsm/kg and a FeNa > 3%, are incorrect. The definition of SIADH, as originally put forth by Bartter and Schwartz, is that in the absence of a hemodyamic stimulus for AVP production, renal disease or endocrine disorder, that the urine is less than maximally dilute (> 100 mOsm/kg) and that urinary sodium excretion increases in response to fluid volume or sodium intake.[2] Under most circumstances patients with SIADH have a urine osmolality less than 600 mOsm/kg.[3] SIADH is one of the most common causes of hospital- acquired hyponatremia and has been associated with almost every disease state.

Dr. Coulthard’s concern that 0.45% saline should be avoided for fear of developing hypernatremia is unfounded. Severe renal concentrating defects are extremely rare. Congenital nephrogenic diabetes insipidus has an estimated prevalence of only 1 in 250,000 males. Most of these children will present with hypernatremia. Patients with renal disease do have partial renal concentrating defects, but it would be unusual, if not unheard of, for a child with normal renal function to be hyposthenuric ( unable to concentrate the urine to an osmolality similar to plasma). An otherwise healthy child will not develop hypernatremia from the use of 0.9% saline as the kidney will excrete a hypertonic urine and generate free water. We have previously demonstrated that the majority of otherwise healthy ambulatory children excrete a hypertonic urine on a regular basis.[4] On the other hand, virtually every hospitalized child has a potential stimulus for AVP production, and the risk of developing hyponatremia from the administration 0.18% saline is real. A recent prospective study has revealed that hypotonic fluids result in a fall in serum sodium, while isotonic fluids do not produce hypernatremia.[5] Coulthard himself has reported that 0.18% saline and near isotonic Hartman’s solution do produce a fall in serum sodium post-operatively and do not produce hypernatremia.[6] There can be no justification for the routine administration of 0.18% saline, and there is neither physiological basis nor empirical evidence to support the idea that isotonic fluids will produce hypernatremia.

1. Moritz ML, Ayus JC. Prevention of hospital-acquired hyponatremia: a case for using isotonic saline. Pediatrics 2003;111:227-30.

2. Bartter FC, Schwartz WB. The syndrome of inappropriate secretion of antidiuretic hormone. Am J Med 1967;42:790-806.

3. Musch W, Decaux G. Treating the syndrome of inappropriate ADH secretion with isotonic saline. Qjm 1998;91:749-53.

4. Moritz ML. Urine sodium composition in ambulatory healthy children: hypotonic or isotonic? Pediatr Nephrol 2008. Epub ahead of print

5. Yung M, Keeley S. Randomised controlled trial of intravenous maintenance fluids. J Paediatr Child Health 2007;0:0.