We have evaluated the efficacy and safety of granisetron, a selective 5-hydpoxytryptamine type-3 receptor antagonist, for the prevention of postoperative nausea and vomiting (PONV) in women undergoing thyroidectomy. In a prospective, randomized, placebo-controlled, double-blind study, 100 ASA patients, aged 30-57 yr; received placebo or granisetron at three different doses (20, 40 or 100 mu g kg(-1)) (n=25 each), i.v., immediately before induction of anaesthesia. A standard general anaesthetic technique was used. A complete response, defined as no PONV and no need for another rescue antiemetic during the first 3 h after anaesthesia, was seen in 36%, 44%, 92% and 92% of patients who received placebo, granisetron 20 mu g kg(-1), 40 mu g kg(-1) and 100 mu g kg(-1), respectively; corresponding values during the next 21 h after anaesthesia were 40%, 44%, 88% and 88% (P<0.05; overall Fisher's exact probability test). There was no clinically important adverse events in any group. We conclude that granisetron 40 mu g kg(-1) was an effective antiemetic for the prevention of PONV after thyroidectomy. Increasing the dose to 100 mu g kg(-1) provided no further benefit.