Page 87 - 15Diarrhoeaandvomiting

Diarrhoea and vomiting caused by gastroenteritis in children under 5 years

with severe dehydration were included while five other trials included malnourished children.
Nine trials reported allocation concealment that was adequate and six were double-blinded.
Loss to follow-up was less than 10% of the randomised participants in all trials. The protocol of
this review had initially defined reduced osmolarity as < 250 mOsm/l but during the course of
reviewing this limit was increased to 270 mOsm/l or less since some trials had used this higher
limit of definition. Osmolarity of the control group was also increased from 311 to 331 mmol/l
to include two additional trials. Nevertheless, in both the trials the concentration of sodium and
glucose was similar to the WHO-recommended ORS solution. Since stool output was measured
in various ways using different units in the RCTs, their results were pooled and expressed as
standardised mean difference.
Out of 14 trials, 11 reported the need for unscheduled IVT. Three trials reported that none of the
children required an IV infusion and hence odds ratios could not be calculated. Results from the
meta-analysis of the other eight trials (n = 1996) showed a significant reduction in the need for
additional IVT for children receiving the low-osmolarity ORS solution compared with children
treated with the WHO-recommended high-osmolarity ORS solution (OR 0.59; 95% CI 0.45
to 0.79). Sensitivity analysis conducted with studies where allocation concealment was clearly
described as adequate suggested little difference in the result for the primary outcome. Eleven
trials (n = 1776) measured stool output during the rehydration period and the pooled results
showed a significant reduction in stool output with the low-osmolarity ORS solution (SMD −0.23;
95% CI −0.33 to −0.14). Hyponatraemia and vomiting during rehydration were reported in six
trials each. Children treated with the reduced osmolarity ORS solution showed a lower tendency
for vomiting (OR 0.71; 95% CI 0.55 to 0.92) compared with the WHO ORS solution group, but
no statistically significant difference was observed for the presence of hyponatraemia (OR 1.44;
95% CI 0.93 to 2.24). There was no evidence of statistical heterogeneity for any of the results.
In the second systematic review from the USA, the effectiveness of ORT was evaluated in
90
comparison with IVT among well-nourished children with gastroenteritis in developed countries,
and this was followed by a comparison between high-sodium ORS solution and low-sodium
ORS solution. Trials were included if they were published in English, conducted in populations
of well-nourished children during the late 1970s through to the early 1990s and included more
than ten patients. A total of 13 trials were included in this review and all were conducted in
the USA or Canada – six RCTs of ORT with IVT arms and seven RCTs without IVT arms (i.e.
comparing oral solutions with differing sodium content). The age of the study population ranged
from 3 months to 3 years but one study included children aged 1 month to 14 years. Clinical
efficacy was defined as the success of ORT in rehydrating children with gastroenteritis within
12–24 hours of starting treatment, while failure was defined as the need to use IVT for rehydration.
High sodium content was defined as a sodium level of 90 mmol/l, medium as 50–75 mmol/l and
low as 26–45 mmol/l. Safety was measured by the relative incidence of hypernatraemia (serum
sodium level > 146 mmol/l) and hyponatraemia (serum sodium level < 132 mmol/l) induced by
the treatment. [EL = 1+]
Altogether, eight trials (one RCT with an IVT arm and seven trials without an IVT arm) had compared
ORT solutions of differing sodium content and their results are reported in this section. The high-
sodium formula had the lowest failure rate among the three groups at 1.9% (95% CI 0% to 5.4%),
while the low-sodium group had a failure rate of 3.6% (95% CI 0% to 7.3%) and the medium-
sodium group a failure rate of 5% (95% CI 1.9% to 8.1%). However, there were no statistically
significant differences in the failure rates of the three groups treated with high-, medium- and
low-sodium ORS solution. Only one trial with an IVT arm gave information on the cases of
hypo- and hypernatraemia. It reported three cases of hyponatraemia that corrected to normal
within 24 hours of treatment. Another trial with no IVT arm reported one case of hyponatraemia
in the high-sodium group and six cases each in the medium- and low-sodium groups. Estimates
of effect could not be calculated for incidences of hyponatraemia and hypernatraemia because
the total numbers of individuals in each group were not available. Moreover, there were no
statistically significant differences between the high- and low-sodium ORS solution for other
outcomes (weight gain, volume, frequency and duration of diarrhoea, or length of hospital stay).
91
A Cochrane review compared rice-based ORS solution (50–80 g/l of rice powder) to the glucose-
based WHO ORS solution (20 g/l of glucose) for the treatment of diarrhoea. Trials were included
only if the rice-based ORS solution was made by replacing glucose in the standard WHO ORS

62