Postoperative nausea and vomiting (PONV) is a commonly observed adverse effect of general anesthesia. Recently, ondansetron, a new serotonin3 (5-hydroxytryptamine3) receptor antagonist was shown to be effective in the prophylaxis and prevention of chemotherapy-induced nausea and vomiting in children and adults as well as of PONV in adults. The aim of the current study was to evaluate the capacity of ondansetron to prevent PONV in pediatric patients. Two hundred children (132 boys and 68 girls) 2-10 yr of age received general inhalational anesthesia for surgical procedures (the extremities; ear, nose, and throat; inguinal hernia and phimosis; and dentistry) of an expected duration of less than 90 min. This study was divided into two phases: prophylaxis and rescue treatment. For prophylaxis, patients were randomly assigned to two groups: one group received an intravenous injection of 0.1 mg/kg ondansetron, and the other group received a placebo before surgical incision under double-blind conditions. For rescue treatment, only placebo patients were included; as a rescue medication they received an intravenous injection of 0.1 mg/kg ondansetron or 0.02 mg/kg droperidol according to a prestudy randomization under double-blind conditions. Incidence and severity of PONV (PONV score 0 = no nausea and no retching; 1 = complaining of sickness and retching; 2 = vomiting one or two time in 30 min; 3 = vomiting more than two times in 30 min) was recorded over a 4-h period in the postanesthesia care unit. Within 72 h of the procedure, a follow-up nurse interviewed the parents for late-onset nausea in the children. With regard to prophylaxis, 10% of patients receiving ondansetron had PONV during the 4-h observation period versus 40% of those receiving placebo (P < 0.001). The incidence of vomiting alone (PONV score > or = 2) was 5% and 25%, respectively (P < 0.001). There were no significant differences between ondansetron and droperidol in the treatment of PONV. However, at the end of the 4-h period, ondansetron patients were less sedated than were patients who had received droperidol (P < 0.01). Interviews with parents could be performed for 143 of 200 children (76 ondansetron and 67 placebo). Twenty-four children (15 ondansetron and 9 placebo) showed late-onset PONV after the 4-h observation period but within 24 h of the procedure (19.7% vs. 13.4%; P not significant). Ondansetron is effective in the prevention of PONV in pediatric patients for the first 4 h after general anesthesia. Lower sedation scores with ondansetron compared with droperidol may be an advantage, especially in ambulatory surgery. However, the incidence of late-onset PONV (> 4-24 h) was not influenced by prophylactic treatment with one dose of ondansetron preoperatively.