Page 122 - 15Diarrhoeaandvomiting

Antibiotic therapy

groups and results were presented for 32 antibiotic and 33 placebo recipients. There were no
statistically significant differences between the two groups in the mean duration of abnormal stool
frequency, vomiting, dehydration or fever. However, the erythromycin group had a statistically
significantly shorter mean duration of abnormal stool consistency compared with the placebo
group (WMD −0.80 days; 95% CI −1.46 to −0.14 days). [EL = 1+]

154
A third RCT, conducted in Mexico, with three treatment arms compared the effects of
furazolidone, trimethoprim/sulfamethoxazole and no antibiotic treatment for acute invasive
diarrhoea in children. Patients aged 2–59 months brought to hospital and seen in outpatients
with three or more watery stools in the previous 24 hours, with up to 5 days of diarrhoea prior
to admission, and with presence of polymorphonuclear (PMN) leucocytes and blood in stool
(n = 125) were included in the study. Patients who had received antimicrobials or antidiarrhoeal
drugs in the previous 48 hours, who had amoeba in their stools, who had any concomitant
disease or who had allergy or intolerance to the study drugs were excluded. Following a complete
physical examination and submission of a stool specimen, participants were randomised to
receive 7.5 mg/kg per day furazolidone in four equal doses a day for 5 days (n = 42), 8 mg/kg
per day trimethoprim + 40 mg/kg per day sulfamethoxazole in two equal doses a day for 5 days
(n = 52) or no treatment (n = 24). Participants were followed up with daily visits as outpatients
to hospital, clinical assessment at day 3 and stool samples taken at days 1 and 6. Treatment
success for participants with an identified pathogen was defined as clinical cure (absence of
diarrhoea and alleviation of all symptoms) at day 3 and bacteriological cure (negative stool
culture) at day 6. For patients with negative culture, treatment success was defined as clinical
cure (absence of diarrhoea and alleviation of symptoms) at day 3. The methods of randomisation
were not reported and it was unclear as to whether patients or outcome assessors were blinded
to treatment. [EL = 1−]
At baseline, patients were similar for age, sex, weight, height, body temperature and mean stools
passed per day. However, those children receiving furazolidone had fewer days with diarrhoea
prior to recruitment compared with patients receiving either trimethoprim/sulfamethoxazole or
no antibiotic treatment (P < 0.02). A statistically significantly higher number of patients (who took
antibiotics) were clinically cured by day 3 compared with the no antibiotic group (furazolidone
RR 1.93; 95% CI 1.21 to 3.09, trimethoprim/sulfamethoxazole RR 1.82; 95% CI 1.13 to 2.92, and
for both antibiotics together RR 1.87; 95% CI 1.18 to 2.98). Similar results were seen for clinical
cure rate by day 6 (furazolidone RR 2.78; 95% CI 1.25 to 6.19, trimethoprim/sulfamethoxazole
RR 3.05; 95% CI 1.38 to 6.72, and for both antibiotics together RR 2.92; 95% CI 1.33 to 6.39).
However, among those patients who had negative stool cultures, there were no statistically
significant differences in the proportion of patients who had been clinically cured at day 3, for
either furazolidone or trimethoprim/sulfamethoxazole individually or for both antibiotics together
compared with no antibiotic treatment. For patients with positive stool cultures, bacteriological
cure at day 6 was only statistically significantly different from placebo when data for antibiotics
were combined (RR 2.33; 95% CI 1.04 to 5.25). No statistically significant differences were found
for furazolidone (RR 1.76; 95% CI 0.76 to 4.12) or trimethoprim/sulfamethoxazole (RR 1.97;
95% CI 0.85 to 4.56) alone compared with placebo.
153
Another RCT from Mexico recruited children aged 3–84 months seen in hospital with diarrhoea
into a treatment trial of trimethoprim/sulfamethoxazole compared with placebo. Participants
had passed three or more unformed stools in the previous 24 hours, had less than 72 hours’
duration of diarrhoea, no antibiotic treatment in the previous 7 days and were not severely
dehydrated (n = 141) and were randomised into two treatment groups to receive 10 mg/kg per day
trimethoprim + 50 mg/kg per day sulfamethoxazole oral suspension in two divided doses a day for
5 days (n = 73) or placebo oral suspension (n = 68). Daily assessments were made throughout the
duration of treatment and once more at 2 weeks. Although details of the randomisation process
were not reported, patients and outcome assessors were blind to treatment allocation. The groups
were similar at study entry for age, pre-study diarrhoea duration, mean stool passage at 24 hours
pre-therapy, fever, dehydration status and faecal leucocytes. Fifty of 141 participants had body
weight under the third percentile for age according to Mexican standard criteria. [EL = 1−]

The mean time to last diarrhoeal stool was statistically significantly shorter with antibiotic use
compared with placebo in all patients (58.2 versus 75.5 hours; P = 0.021), those with fever
(58.2 versus 75.5 hours; P = 0.021) and those with faecal leucocytes (>3/HPF) (57.7 versus

97