Page 137 - 15Diarrhoeaandvomiting

Diarrhoea and vomiting caused by gastroenteritis in children under 5 years

8.2.2.1 Racecadotril

Two trials compared the effect of racecadotril with placebo. 169,170 Both trials had two treatment
arms and, in total, data from 307 participants (aged 3 months to 4 years) were collected.
One trial (n = 135) conducted in Peru 169 examined the effect of racecadotril compared with
placebo in boys aged between 3 and 35 months admitted to hospital for dehydration. The included
participants had passed watery diarrhoea for 5 days or less, had passed three or more diarrhoeic
stools in the 24 hours prior to admission and had passed one diarrhoeic stool within 4–6 hours
post admission. Those with blood in the stool, severe dehydration (inability to drink because of
drowsiness) or any serious concomitant illness were excluded. Racecadotril (1.5 mg/kg) (n = 68)
or placebo (n = 67) was randomly administered as a powder every 8 hours for 5 days or until
diarrhoea stopped if earlier. Oral rehydration was given as needed to all participants. No details
were provided about sequence generation, allocation concealment or blinding of assessors. The
recorded outcomes included stool output, duration of diarrhoea and overall cure rate measured
at 5 days. [EL = 1+]
Both groups were broadly comparable at baseline. Seventeen percent of all participants were
lost to follow-up. A statistically significant difference was reported in the mean 48 hour stool
output favouring racecadotril over placebo for all participants and for both the rotavirus positive
and negative groups. For all 135 participants, the mean stool output in the first 48 hours was 92
± 12 g/kg in the racecadotril group (n = 68) compared with 170 ± 15 g/kg in the placebo group
(n = 67) (P < 0.001). A statistically significant reduction in mean stool output was also observed in
the rotavirus-positive participants (n = 73). In the racecadotril group (n = 34), the mean stool output
in the first 48 hours was 105 ± 17 g/kg and 195 ± 20 g/kg in the placebo group (n = 39) (P < 0.001).
The authors stated that in the rotavirus-negative subgroup (n = 62) there was a statistically significant
reduction in the mean 48 hour stool output in participants receiving racecadotril compared with
those receiving placebo (31% lower in the racecadotril group; 95% CI 16% to 46%; P < 0.001).
For all participants, the mean hourly rate of stool production in first 48 hours was statistically
significantly lower in the racecadotril group (1.8 ± 0.2 g/kg/hour) compared with the placebo
group (3.1 ± 0.3 g/kg/hour) (P < 0.001). The mean total stool output before recovery was statistically
significantly lower in the racecadotril group (157 ± 27 g/kg) compared with the placebo group
(331 ± 39 g/kg) (P < 0.001). This was also observed in the rotavirus-positive participants (n = 73): in
the racecadotril group (n = 34) the mean stool output was 174 ± 36 g/kg and in the placebo group
(n = 39) it was 397 ± 37 g/kg (P < 0.001). Although no further details were provided, the authors
stated that in the rotavirus-negative subgroup (n = 62), there was a statistically significant reduction
in the mean stool output before recovery in participants receiving racecadotril compared with
those receiving placebo (37% lower in the racecadotril group; 95% CI 20% to 56%; P < 0.001).
The median duration of diarrhoea was shorter for the racecadotril group than the placebo group in
both the rotavirus subgroups. In both subgroups, the median duration of diarrhoea was 28 hours
for the racecadotril group. However, in the placebo group the rotavirus-positive participants had
a median duration of diarrhoea of 72 hours compared with 52 hours in the rotavirus-negative
participants (P < 0.001). In all participants, at 5 days, 57 of the racecadotril group (n = 68) were
cured of diarrhoea (passing of two consecutive formed stools or not having passed a stool for
12 hours) compared with 44 cured participants in the placebo group (n = 67). This difference was
statistically significant (RR 1.28; 95% CI 1.04 to 1.56; P = 0.02).
A multicentre RCT (n = 172), 170 conducted in 13 centres in France, examined the effect of
racecadotril compared with placebo in children of both sexes aged between 3 months and
4 years hospitalised for severe acute diarrhoea. Included participants had passed watery
diarrhoea (three watery stools per day or more) for less than 72 hours’ duration and had passed
one watery stool post admission to hospital. Children were excluded if they had chronic
diarrhoea, a weight-for-age deficit of 20% or more of the US National Center for Health Statistics
(NCHS) standard, a systemic illness or any antibiotic, antidiarrhoeal drug or acetylsalicylic acid
usage in the preceding 48 hours. Racecadotril (1.5 mg/kg) (n = 89) or placebo (n = 83) was
randomly administered as a powder three times daily for 5 days or until diarrhoea stopped if
earlier. Rehydration was administered orally or by gastric tube without restriction. No details
were provided about sequence generation, allocation concealment or blinding of assessors and
there were considerable losses in data collection and follow-up (28%). Measures of stool output
were presented as both a full data set (n = 168) and per-protocol results (n = 121). [EL = 1−]

112