The effects of clonidine as a preanesthetic medication were compared with diazepam on clinical courses of sevoflurane anesthesia in 22 patients undergoing upper abdominal surgery. The patients were divided into two groups of 11 patients each according to preanesthetic medication: atropine 0.5 mg i.m. plus clonidine 0.3 mg p.o., or atropine 0.5 mg i.m. plus diazepam 10 mg p.o. 60-90 min prior to induction of anesthesia. Anesthesia was induced with fentanyl and thiopental, and was maintained with sevoflurane, 0.5%-1.5%, nitrous oxide and oxygen, supplemented with fentanyl, 0.5 μg·kg·hr. While only one patient needed a vasodilator in the clonidine group for treatment of hypertension, seven patients needed it in the diazepam group. Pain score after extubation was higher in the diazepam group than in the clonidine group. The time when patients responded to verbal command after discontinuation of anesthetics was similar in both groups. Therefore, clonidine pretreatment was useful for sevoflurane anesthesia in upper abdominal surgery.