Page 105 - 15Diarrhoeaandvomiting

Diarrhoea and vomiting caused by gastroenteritis in children under 5 years

stated to be statistically significant (t = 2.14; P < 0.004). The children in the former two groups
gained on average 140 g more than those in the latter groups. However, the authors did not report
statistically significant differences in the duration of diarrhoea or therapeutic failure rates (defined
as recurring dehydration, worsening electrolyte abnormalities or prolonged severe diarrhoea)
between the four groups. [EL = 1−]

Another RCT, from Pakistan, recruited 69 boys aged between 9 and 48 months with acute
120
watery diarrhoea (duration less than 3 days), moderate or severe dehydration, no previous
antibiotic treatment and no complication other than those directly related to dehydration, and
who were weaned from mother’s milk. Criteria for exclusion were severe systemic illness, severe
malnutrition, oedema or fever more than 101 °F. After initial rehydration with ORS solution or
IVT (duration not given), children were randomly allocated (using a random number table) to the
two groups:
• Group A (late feeding group n = 33), where children received only ORS solution for 24 hours
followed by khitchri (culturally acceptable food made from rice, legumes and cottonseed oil)
and half-strength cow’s milk formula
• Group B (n = 36), where children received khitchri and half-strength cow’s milk formula
along with ORS solution immediately after rehydration.
The admission characteristics of the two groups were similar with regard to age, weight, vomiting,
purging rate, dehydration status and nutritional status. No statistically significant differences were
seen between the two groups for weight gain (at 24 hours and 72 hours post rehydration), mean
stool output or the number of stools. There was also no difference between the two groups
regarding number of treatment failures (children started on IVT). [EL = 1−]
41
A quasi-randomised study carried out in Israel in a primary care unit involved 90 infants aged
1–12 months with acute watery diarrhoea (duration up to 7 days) and mild dehydration. Excluded
were babies younger than 30 days, children born prematurely, those receiving antibiotic therapy,
those with moderate to severe dehydration and those whose parents refused to participate in the
study. Allocation to the early feeding (n = 53) or the late feeding (n = 37) group was done by
flipping a coin and children in both the groups were re-fed after an initial rehydration period with
WHO ORS solution. In the early feeding group, children were given ORS solution for 6 hours
(50 ml/kg), following which parents were advised to continue the same feeding that was being
given prior to presentation and alternate it with ORS solution (75 ml/kg every 18 hours). In
the late feeding group, only ORS solution was given for the initial 24 hours (200 ml/kg per
day) and then feeding introduced. The two groups were similar regarding baseline demographic
characteristics and clinical features on presentation. The outcomes were assessed at 24 hours
and at 2 weeks but there was a high drop-out rate (11% at 24 hours and 30% at 2 weeks). There
were no statistically significant differences between the two groups for any of the outcomes –
percentage weight gain, state of dehydration, duration of diarrhoea or hospital admissions, at
either 24 hours or at 2 weeks. [EL = 1−]
In another quasi-randomised trial, from Romania, 121 122 infants (aged 1–12 months) with acute
diarrhoea (duration up to 5 days) and without signs of severe dehydration were recruited for
the study. Infants with frequent episodes of vomiting, ileus or severe dehydration/shock were
excluded. Children were allocated to the early feeding (n = 73) or late feeding (n = 49) group
depending on the day of the week they were examined (odd or even). In the early feeding
group, a non-restrictive diet was used, that is, in breastfed infants breastfeeding was continued
or in non-breastfed infants a feeding regimen adapted to age was given after 3–6 hours of initial
rehydration with ORS solution or rice water. The feeding regimen used prior to the onset of illness
was reached within 2–3 days in this group. In the late feeding group, breastfeeding or formula
feeding was discontinued for 24–36 hours and only ORS solution given for the first 6–12 hours.
In the next 24 hours, carrot soup and rice water were introduced and gradually normal feeds
were introduced so that the normal feeding regimen (prior to onset of illness) was resumed within
4–6 days. The baseline demographic characteristics, including the proportion of children with
mild/moderate dehydration and proportion with pathogens identified in stool examination, were
similar between the two groups. The mean percentage weight gain in the early feeding group
was significantly higher compared with the late feeding group at 7 days (+1.2% ± 1.1% versus
−0.01% ± 0.9%; P = 0.01). Moreover, the proportion of infants with weight loss (compared with
the pre-illness weight) was significantly lower in the early feeding group (6.2% versus 37.2%;

80