Page 146 - 15Diarrhoeaandvomiting

Other therapies

stool frequency (four RCTs) showed no statistically significant differences between the zinc and
placebo groups, although two studies presenting results for each of these outcomes did report
statistically significant improvement with zinc administration. When data from four trials were
combined this showed a statistically significantly increased number of vomiting events among
children supplemented with zinc when compared with children in the control group. There
was significant heterogeneity in the meta-analyses undertaken in this review. Adjustment for
nutritional, geographical and treatment differences did not affect the significance of the findings,
suggesting that no one single variable explained the heterogeneity found alone. The applicability
of these results might therefore be limited to those settings with similar population characteristics.

8.3.2 Vitamin A
Evidence overview

Three trials were identified for this review. 187–189 Two of these were conducted in Bangladesh 187,188
189
and one in Turkey. All three had two treatment arms. In total, data from 286 children (aged
6 months to 7 years) suffering from acute diarrhoea were collected across the three studies.
Follow-up, treatment protocol and definition criteria for inclusion of the children with acute
diarrhoea varied between the three studies. The outcomes considered were duration of the
disease, stool output, vomiting, clinical cure, bacteriological cure and treatment failure.
Two studies compared the effects of 200 000 iu vitamin A with placebo in children with
diarrhoea. 187,188.
One RCT (n = 83) 187 examined the therapeutic effect of vitamin A in children suffering from
acute diarrhoea. The participants were boys aged between 1 and 5 years presenting with watery
non-cholera diarrhoea of less than 48 hours’ duration. The study compared the administration
of a single oral dose of 200 000 iu vitamin A plus 25 iu vitamin E with the administration of
placebo, which consisted on 25 iu vitamin E. The methods of randomisation were adequate.
Allocation concealment details were not given, although the patients and outcome assessors
were stated to be blinded to treatment allocation. Comparability of the groups at study entry was
adequate in all the studies and the proportion of patients randomised but lost to follow-up was
under 20%. [EL = 1+]

Figure 8.7 Comparison of the effect of zinc versus placebo on the proportion of children who vomited

121