Page 86 - 15Diarrhoeaandvomiting

Fluid management

GDG translation from evidence to recommendation

Although there were no clinical trials on the effectiveness of fluids other than ORS solution in the
treatment of dehydration, the GDG considered that the composition of such fluids was generally
inappropriate. In dehydration due to gastroenteritis, both water and electrolyte replacement is
essential, and non-ORS solution fluids do not usually contain appropriate constituents. ORS
solution was considered the appropriate fluid for oral rehydration.

Clinical question
What is the most effective composition of ORS solution?
ORS solution has been manufactured using a range of constituents in differing concentrations.
Various organic solutes have been included such as glucose, starch and amino acids. Sodium
chloride has been used in varying concentrations. Other non-essential constituents, including
potassium, bicarbonate and acetate are often included. Much research has been carried out
to evaluate the effectiveness and safety of these various solutions. Two key areas of research
have focused on the optimal sodium/osmolar concentration in ORS solution and on the relative
efficacies of glucose versus rice starch as the organic constituent in ORS solution.
The composition of the original WHO ORS solution (glucose 111, sodium 90, potassium 20,
chloride 80 and bicarbonate 30, all in mmol/l) was selected to allow for use of a single solution
that would effectively treat dehydration secondary to diarrhoea caused by various infectious
19
agents and resulting in varying degrees of electrolyte loss. However, in developed countries,
viral gastroenteritis is common and is associated with less severe salt losses, and so there was
concern that the sodium content of the original WHO ORS solution might be excessive. From
88
the 1970s, efforts focused on improving the efficacy of ORS solution by altering its composition.
It was found that solutions with higher concentrations of co-transporters (such as sugars) and
higher osmolarity decreased rather than increased intestinal sodium and water absorption.
Additionally, hypernatraemia was reported with their use. The current formulation WHO ORS
solution adopted in 2002 (glucose 75, sodium 75, potassium 20, chloride 65 and citrate 10, all
in mmol/l) preserves the 1 : 1 molar ratio of sodium to glucose that is critical for efficient co-
transport of sodium. It has a reduced osmolar load (245 mOsm/l) compared with the original
formulation (311 mOsm/l). It also has a longer pre-mixed shelf life owing to its citrate content.
The evidence searches on this question were limited to include only those studies that compared
the effectiveness of high-osmolarity/low-sodium ORS solution with low-osmolarity/low-sodium
ORS solution or the glucose-based ORS solution with the rice-based ORS solution. Evidence on
other types of ORS solution using different carbohydrate substrates or organic substitutes such as
cereals or amino acids was not reviewed in this section since these products are not available in
the UK and are not currently recommended by the WHO.

Evidence overview
Three systematic reviews have been included – two 89,90 provided evidence relating to effectiveness
of low-sodium/low-osmolarity ORS solution versus high-sodium/high-osmolarity ORS solution,
while the third review compared rice-based ORS solution with the glucose-based ORS solution.
91
One systematic review compared the effectiveness of the previously recommended WHO
89
ORS solution (osmolarity 311 mmol/l with 90 mmol/l of sodium) with reduced osmolarity ORS
solution (osmolarity 250 mmol/l or less with reduced sodium) in the treatment of children with
acute diarrhoea. Only RCTs with adequate randomisation were considered for inclusion while
quasi-randomised trials were excluded. Participants included children with acute diarrhoea (with
history of less than 5 days). The primary outcome of interest was the need for ‘unscheduled’
IV infusion during the course of treatment, while the secondary outcomes were stool output,
vomiting and asymptomatic hyponatraemia (serum sodium < 130 mmol/l) during follow-up.
Results from the various studies were pooled using the fixed effect model. [EL = 1++]

Fourteen RCTs were included in this review and they were conducted in Egypt (two), Bangladesh
(three), Mexico (one), Columbia (one), India (three), Panama (one) and the USA (one). All the
studies recruited children younger than 5 years suffering from acute non-cholera diarrhoea,
with the exception of three trials that did include children with cholera. In five trials, children

61