The efficacy of ondansetron 4 mg was compared with metoclopramide 10 mg for the prevention of post-operative nausea and vomiting in patients after major gynaecological abdominal surgery. Anaesthesia was standardized using thiopentone, atracurium and methadone for induction followed by isoflurane in nitrous oxide-oxygen mixture. Fifty patients were randomized in a double-blind fashion to either receive intravenous (i.v.) ondansetron 4 mg or metoclopramide 10 mg during closure of the pelvic peritoneum. The incidence and frequency of vomiting, and the incidence of severe nausea was recorded for 24 h after surgery. One patient was excluded because of respiratory depression. In the first 4 h after surgery, five patients (20%) in the ondansetron group (n = 25) and eight patients (33%) in the metoclopramide group (n = 24) vomited, whereas at 4-12 h, this increased to 11 patients (44%) and nine patients (37.5%) respectively. The incidence was 52 and 37.5% respectively in the subsequent 12-24 h. The highest incidence of nausea was in the first 4 h after surgery, being 56 and 37.5% in the ondansetron and the metoclopramide groups respectively. This decreased to less than 25% in both groups in the 12-24 h period. Ondansetron 4 mg and metoclopramide 10 mg had similar but short lasting efficacy for the prevention of vomiting in patients who received continued opioid analgesia after major gynaecological surgery.