Page 106 - 15Diarrhoeaandvomiting

Nutritional management

P < 0.01), and a significantly higher number of children in this group gained weight at 7 days
compared with the late feeding group (76.6% versus 32.6%; P = 0.01). However, there were no
statistically significant differences between the two groups regarding the frequency of stools per
day, stool output during hospitalisation or duration of diarrhoea. [EL = 1−]

Evidence summary
There is a lack of quality evidence to answer the question of the benefit of early feeding compared
with late feeding. Results from three studies (two RCTs and one quasi-randomised trial) suggested
that weight gain was higher in children who received early feeding (with breast or cow’s milk
or solid foods) along with ORS solution compared with children who were started on these
feeds after 1–3 days of initial rehydration. Two of these three studies found evidence of weight
gain at 7 days after admission, while one study found evidence of weight gain only in the initial
part of study but not at 5 days or 2 weeks after admission. The other two studies did not find
any evidence of weight gain. All the studies reported no differences between the early feeding
group and the late feeding group for the other outcomes – duration of diarrhoea, stool output or
treatment failure.

6.2.2 Reintroduction of milk or other liquids

Evidence overview
Seven studies are included under this section – six RCTs comparing reintroduction of full-strength
feeding with graded re-feeding regimens, while one RCT looked at the effect of introducing juices
in the feeds of children after appropriate rehydration.
An RCT conducted in two hospitals in Guatemala and Brazil 122 recruited 159 boys aged 15 days
to 6 months with acute diarrhoea (duration less than 5 days), no visible blood in the stool and
no clinical signs of severe dehydration. Almost half of the study population was malnourished.
Excluded from the study were infants with severe malnutrition, who were exclusively or mostly
breastfed, who had systemic infections or who had other infections requiring specific additional
treatments. Children with dehydration were initially rehydrated orally with the WHO ORS
solution and then randomly assigned to receive full-strength cow’s milk formula from the start
(Group A, n = 80) or the formula was reintroduced in a graded manner (half-strength for 24 hours
followed by two-thirds-strength for the next 24 hours and then full-strength cow’s milk formula,
Group B, n = 79). Sealed envelopes were used for randomisation and the investigators and
clinical staff were unaware of the group status. The milk formulas were prepared by a dietitian
who was not involved in the clinical management or data collection, and both the formulas were
given in opaque bottles (150 ml/kg per day divided in eight feedings). Maintenance therapy in the
form of ORS solution and plain water was offered to the infants along with the feeds. The baseline
characteristics of the two groups at the time of admission were comparable and outcomes were
assessed on day 5 at the time of discharge. There were no statistically significant differences
between the two groups for any of the outcomes studied: duration of diarrhoea, percentage
weight gain, stool output (frequency and volume), total intake of milk and ORS solution, or
treatment failure rate. However, the mean energy consumption (in kJ/kg) during the first 24 hours
was significantly higher in the group of children given full-strength formula compared with the
other group (310 ± 130 kJ/kg versus 172 ± 67 kJ/kg; P < 0.05). [EL = 1+]

123
An RCT from the UK recruited 62 infants younger than 6 months admitted to a hospital with
acute gastroenteritis (duration less than 7 days) and mild or moderate dehydration. Infants were
excluded if the diagnosis was not thought to be gastroenteritis or if infants were already receiving
low-lactose preparations for presumed lactose intolerance. After rehydration with ORS solution
for 12 hours, the infants were randomly allocated to receive either full-strength re-feeding (full-
strength cow’s milk formula or breast milk, n = 30) or graded re-feeding (one-quarter-strength for
12 hours, then half-strength for 12 hours followed by full-strength formula, n = 32). No details
were given about the process of randomisation, concealment of allocation or blinding. The two
groups were similar in their baseline characteristics (age, sex, ethnic origin, weight, duration of
symptoms and treatment received before admission). A total of 42% of the infants (26/62) had
recurrence of diarrhoea within 7 days of re-feeding but there were no statistically significant
differences between the two groups for the incidences of lactose intolerance or recurrence. The

81