Page 83 - 15Diarrhoeaandvomiting

Diarrhoea and vomiting caused by gastroenteritis in children under 5 years

clear. After admission in the hospital and recruitment in the study, the children were randomised
to the oral treatment group or the IV treatment group. The oral treatment protocol consisted of
two phases. In the initial rehydration phase, an electrolyte solution with osmolarity 270 mOsm/l
(sodium 80 mmol/l, potassium 20 mmol/l, bicarbonate 35 mmol/l, chloride 65 mmol/l, glucose
70 mmol/l) was administered by nasogastric tube at a rate of 40 ml/kg per hour (maximum 400 ml/
kg) until clinical signs of dehydration had disappeared. This was followed by a maintenance phase
where another electrolyte solution with the same osmolarity but different electrolyte composition
(sodium 40 mmol/l, potassium 30 mmol/l, bicarbonate 25 mmol/l, chloride 45 mmol/l, glucose
130 mmol/l) was given by bottle or nasogastric tube at a rate of 250 ml/kg per day. Children in the
IVT arm were treated for shock with Ringer’s lactate solution at a rate of 20–30 ml/kg as rapidly
as possible or within 1 hour in those with less severe illness. A second infusion of 20–30 ml/kg
was given if the clinical signs of shock persisted. Thereafter two-thirds of the fluid deficit was
replaced during the first 24 hours of treatment and the remaining one-third during the second
day. Abnormal fluid losses due to severe diarrhoea were replaced in both the groups but the
methods were not clearly defined. Failure to rehydrate was defined as ‘no change in the clinical
status or worsening of the signs of dehydration within first 2 hours of treatment’. In such cases,
ORT was discontinued and IVT commenced. [EL = 1−]

The baseline characteristics of the ORT group (n = 236) were similar to those of the controls
treated with IVT (n = 234). In the ORT group, one child failed to rehydrate while there were
no rehydration failures in the IVT group, and there was statistically no statistically significant
difference in the risk of rehydration failure between the two groups (RR 2.97; 95% CI 0.12 to
72.65). The mean duration of diarrhoea was significantly shorter in the group receiving ORT than
in the group treated with IVT (4.8 versus 5.5 days; MD −0.70 days; 95% CI −1.16 to −0.24 days)
and children in the ORT group had a higher percentage weight gain at discharge compared with
the IVT group. At 24 hours after admission, electrolyte abnormalities were recorded in 14/236
children in the ORT arm and in 29/234 children in the IVT arm. A larger number of children
in the IVT group were hypernatraemic or hyponatraemic compared with the ORT group (12
versus one and 13 versus seven, respectively). Hyperkalaemia occurred in three children in the
IVT group and in five in ORT group. However, none of the observed differences in electrolyte
abnormalities between the two groups were statistically significant.
Vomiting (1–3 episodes during the first 6 hours) was more frequent with IVT than ORT during the
rehydration phase (30% versus 19%; P < 0.001). There were no differences between the groups
in the frequencies of abdominal distension or peri-orbital oedema. There were seven deaths in all
– two in the ORT group and five in the IVT group. All who died had completed rehydration, and
most had normal electrolyte levels. Four who died had a body weight below the 3rd percentile.
Home follow-up was carried out for 172 of the ORT group and 169 of the IVT group, but the
study did not specify the number of re-admitted patients treated with ORT and IVT.
The RCT from Indonesia included 75 children (age range 1 to 59 months) with acute diarrhoea
84
and severe dehydration. Criteria for inclusion were the presence of a palpable and countable
pulse, and absence of abdominal distension and other complications. The authors did not define
their criteria for severe dehydration. Following recruitment, children were randomised to the
ORT or IVT group using predetermined random numbers. The ORT group received WHO-
recommended ORS solution by nasogastric infusion while the IVT group received Ringer’s
lactate solution. In both the groups, fluid administration rates were in accordance with WHO
recommendations (40 ml/kg in the first hour, 30 ml/kg in the second, 20 ml/kg in the third and
20 ml/kg in the fourth hour). However, the definition of ‘rehydration failure’ was not consistent
for the two groups – in the ORT group it was taken as cessation of oral therapy and start of IVT
due to increased frequency of vomiting and diarrhoea within the first 4 hours of treatment, while
in the IVT group it was continuation of IV fluid longer than 4 hours due to excessive vomiting or
seizures. [EL = 1−]
At baseline there were no statistically significant differences between the nasogastric ORT group
(n = 36) and the IVT group (n = 39) in relation to mean body weight, mean frequency or duration
of diarrhoea, or mean frequency or duration of vomiting before admission. In the ORT group,
3/36 children (8.3%) failed to rehydrate and in the IVT group 2/39 children (5.1%) failed to
rehydrate, and this difference was not statistically significant (RR 1.63; 95% CI 0.29 to 9.17).
Two children given ORT and four given IVT experienced a recurrence of dehydration after initial

58