Does oral ondansetron reduce vomiting and the need for intravenous fluids and hospital admission in children presenting with vomiting secondary to gastroenteritis?
| Study | Study group | Study type (level of evidence) | Outcome | Key results | Study weaknesses and comments |
|---|---|---|---|---|---|
| Yilmaz et al,14 Turkey | 109 patients 5 months–8 years old with physician-diagnosed gastroenteritis unable to tolerate oral feeding, with at least four vomits in the previous 6 h and four episodes of diarrhoea in the previous 24 h, and mild-to-moderate dehydration. Randomised to orally dissolving ondansetron 0.2 mg/kg or placebo, then ORT restarted 30 min later. Total of three doses given in 24 h period | Prospective double-blind RCT (level 1b). Intention-to-treat analysis | Primary outcome: frequency of vomiting in 8 h after enrolmentSecondary outcomes: need for intravenous hydration, admission to hospital | 12/55 in treatment group vs 36/54 in placebo had vomiting in first 8 h: RR 0.33 (CI 0.19 to 0.56); NNT 2 (CI 1.6 to 3.5) (p<0.001). 6/55 in treatment group vs 39/54 in placebo group had vomiting over 24 h follow-up period: RR 0.15 (CI 0.07 to 0.33); NNT 2 (CI 1.3 to 2.1) (p<0.001). Total percentage of patients hospitalised and/or who received intravenous rehydration was 5.4% in treatment group and 18.6% in placebo group: RR 0.29 (CI 0.09 to 1.01) (p=0.04) | Small increase in diarrhoea in treatment vs placebo group significant at 24 h (5.0 episodes in treatment group vs 4.3 in placebo; p=0.04), but not at 8 h. Patients only recruited for 2 h on weekday mornings due to restrictions on observation time |
| Roslund et al,6 USA | 106 patients 1–10 years old with acute gastritis or gastroenteritis and mild–moderate dehydration who had failed ORT in A&ER. Randomised to one dose of orally dissolving ondansetron tablets 0.15 mg/kg or placebo, then ORT restarted 30 min later | Prospective double-blind RCT (level 1b). Intention-to-treat analysis | Primary outcome: need for intravenous hydration. Secondary outcomes: admission to hospital, cessation of vomiting | 11/51 in treatment group required intravenous fluids vs 30/55 in placebo group: ARR 0.33; NNT 3 (p<0.001). 3/51 vs 7/55 required admission. 93% in treatment group vs 88% in placebo group had no further episodes of vomiting after discharge | Small increase in diarrhoea in treatment vs placebo group on telephone follow-up (median number of episodes 0 in both groups). Commented that ondansetron preparation used may cause less diarrhoea than syrup preparation |
| Freedman et al,9 North America | 215 patients 6 months–10 years old treated in the emergency department for gastroenteritis or dehydration. Received one dose of orally dissolving ondansetron (2 mg for weight <8–15 kg, 4 mg for 15–30 kg and 8 mg for >30 kg) or placebo, and then started on ORT | Prospective double-blind RCT (level 1b). Intention-to-treat analysis | Primary outcome: proportion who vomited while receiving oral rehydration. Secondary outcomes: number of episodes of vomiting, need for intravenous hydration or admission to hospital | 15/107 in treatment group vs 37/107 in placebo group vomited during ORT: RR 0.40 (CI 0.26 to 0.61) (p<0.001); NNT 5.15/107 in treatment group vs 33/107 in placebo group required intravenous fluids: RR 0.46 (CI 0.26 to 0.79) (p=0.003); NNT 6. Non-significant difference in admission rates | Children in ondansetron group had more episodes of diarrhoea during ORT than placebo group (1.4 vs 0.5; p<0.001), even after adjustment for the number of episodes occurring before arrival. The revisit rate by 1 week after intervention was 19% in the ondansetron group and 22% in the placebo group |
| Ramsook et al,10 North America | 145 children 6 months–12 years old with at least five episodes of vomiting in preceding 24 h with or without diarrhoea, and clinical diagnosis of gastroenteritis. Ondansetron syrup (1.6 mg for patients aged 6 months–1 year, 3.2 mg aged 1–3 years, and 4 mg aged 4–12 years) or placebo was given every 8 h for up to 2 days. ORT was started 15 min after the first dose of ondansetron or placebo | Prospective double-blind RCT (level 1b). Intention-to-treat analysis | Primary outcomes: vomiting during the 48 h after enrolment and need for intravenous hydration. Secondary outcomes: admission to hospital and frequency of diarrhoea | 64/74 in treatment group vs 46/71 in placebo group had no further vomiting in ED: ARR 0.21; NNT 5 (p=0.004). 8% of treatment group vs 23% of placebo group required intravenous fluids (p=0.015). 2/74 in treatment group vs 11/71 in placebo group required admission (p=0.007)Increase in diarrhoea in treatment group significant after discharge: ARI 0.05; NNH 21 (p=0.02) | The study was supported in part by a grant from GlaxoWellcome Research and Development who manufacture ondansetron |
All CIs shown in table are for 95%.
A&E, Accident and Emergency; ARI, absolute risk increase; ARR, absolute risk reduction; NNH, number needed to harm; NNT, number needed to treat; ORT, oral rehydration therapy; RCT, randomised controlled trial.