One hundred and thirty-one children with a median age of 5 years were administered Navoban (tropisetron), a selective antagonist of the serotonin receptor (5-HT3), dosed once daily at 0.2 mg/kg (with a maximum of 5 mg daily) in a study aimed at evaluating the prevention of nausea and vomiting induced by anti-cancer chemotherapy. The most common malignancy (in 49% of patients) was acute lymphocytic leukaemia. Patients received Navoban during one or more courses of emetogenic chemotherapy for a total of 455 courses administered intravenously or intravenously and intrathecally (IV + IT). Most patients (89%) had already received cytotoxic chemotherapy before enrollment for the trial. On Day 1, Navoban was administered slowly and intravenously as a single dose before the start of chemotherapy, or by mouth as a single daily dose on subsequent days (median treatment duration = 5 days). On the first 5 days of each course of chemotherapy, response to Navoban per 24-hour period was graded as: complete (absence of both nausea and vomiting), partial (1-4 vomits and/or less than 5 hours of nausea), or failure (more than 4 vomits and/or at least 5 hours of nausea). Ninety-six per cent of the intravenous chemotherapy group and 97% of the IV + IT chemotherapy group had a complete (70% and 55% respectively) or partial (26% and 42% respectively) response during the first 24-hour period of the first course in which Navoban was used. The second and subsequent courses yielded similar percentages. Delayed emesis was observed mainly during those courses employing the most emetogenic chemotherapy.(ABSTRACT TRUNCATED AT 250 WORDS)