BACKGROUND Dexmedetomidine, an α(2)-receptor agonist, provides sedation, analgesia, and anxiolytic effects, and these properties make it a potentially useful anesthetic premedication. In this study, we compared the effects of intranasal dexmedetomidine and midazolam on mask induction and preoperative sedation in pediatric patients. METHODS Ninety children classified as ASA physical status I, aged between 2 and 9, who were scheduled to undergo an elective adenotonsillectomy, were enrolled for a prospective, randomized, and double-blind controlled trial. All of the children received intranasal medication approximately 45-60 min before the induction of anesthesia. Group M (n = 45) received 0.2 mg·kg(-1) of intranasal midazolam, and Group D (n = 45) received 1 μg·kg(-1) of intranasal dexmedetomidine. All of the patients were anesthetized with nitrous oxide, oxygen, and sevoflurane, administered via a face mask. The primary end point was satisfactory mask induction, and the secondary end points included satisfactory sedation upon separation from parents, hemodynamic change, postoperative analgesia, and agitation score at emergence. RESULTS Satisfactory mask induction was achieved by 82.2% of Group M and 60% of Group D (P = 0.01). There was no evidence of a difference between the groups in either sedation score (P = 0.36) or anxiety score (P = 0.56) upon separation from parents. The number of patients who required postoperative analgesia was higher in the midazolam group (P = 0.045). CONCLUSIONS Intranasal dexmedetomidine and midazolam are equally effective in decreasing anxiety upon separation from parents; however, midazolam is superior in providing satisfactory conditions during mask induction.