We assessed the safety and efficacy of oral transmucosal fentanyl citrate (Fentanyl Oralet; Abbott Laboratories, Abbott Park, IL), administered preoperatively to provide both preoperative sedation and postoperative analgesia, in a randomized, double-blind, placebo-controlled study in 40 children, 2-10 yr of age, scheduled for tonsillectomy. In the preoperative holding area, one group (Group O) received Fentanyl Oralet (fentanyl 10-15 micrograms/kg), and the other (Group IV) received only the candy matrix. Patients in Group O received an i.v. injection of saline, and those in Group IV received an i.v. injection of fentanyl (2 micrograms/kg) after removal of the first tonsil. Except for the opioid, patients received a standard anesthetic. Preoperative sedation and cooperation were assessed. Postoperative pain was evaluated using an objective pain scale. Patients in Group O were more sedated but no more cooperative at the induction of anesthesia compared with those in Group IV. No patient vomited preoperatively or experienced preoperative or postoperative desaturation. Time to postanesthesia care unit (PACU) discharge was not different between groups. There was no significant difference in the number of patients requiring morphine in the PACU (6 of 21 in Group O versus 10 of 19 in Group IV). Plasma fentanyl concentrations were not a reliable indicator of the need for postoperative morphine. Among the patients who required morphine postoperatively, there was an 11-fold variation in plasma fentanyl concentrations at the time of morphine administration. Derived pharmacokinetic parameters were similar to those previously reported in children; bioavailability of the fentanyl in Fentanyl Oralet was 0.33. We conclude that premedication with Fentanyl Oralet did not differ with i.v. fentanyl in regard to the induction of anesthesia and postoperative analgesia. CONCLUSIONS In this double-blind, randomized study, we studied the efficacy of Fentanyl Oralet (10-15 micrograms/kg) preoperatively for providing postoperative analgesia in children undergoing tonsillectomy. We found no incidence of preoperative desaturation or vomiting in any patient. This is in contrast to other studies, in which there was a longer time interval between Fentanyl Oralet completion and induction of anesthesia. The bio-availability of the fentanyl in Fentanyl Oralet was estimated to be 33%, which is less than that reported in adults (approximately 50%). There was no difference in postoperative opioid requirements between patients who received 2 micrograms/kg of fentanyl i.v. and those who received Fentanyl Oralet.