In adults, intraoperative administration of tramadol could result in earlier recovery and less sedation than morphine. In this controlled, randomized, double-blind study, we investigated whether an intraoperative initial dose of tramadol could cause more rapid awakening from general anesthesia, less sedation, and earlier tracheal extubation than morphine in children during the immediate postoperative period. Forty children aged 1-6 yr, scheduled for atrial or ventricular septal defect repair and tracheal extubation in the pediatric intensive care unit, were randomly allocated to receive morphine, initial dose 0.2 mg/kg, or tramadol 2 mg/kg given at the end of sternal closure, followed by nurse-controlled analgesia (bolus 0.02 mg/kg of morphine and 0.2 mg/kg of tramadol) with background infusions (0.015 mg x kg(-1) x h(-1) for morphine and 0.15 mg x kg(-1) x h(-1) for tramadol). Postoperatively, children receiving tramadol had earlier awakening from general anesthesia (P = 0.02) and were less sedated at 1 and 2 h postoperatively (P = 0.03 and P = 0.01, respectively). Tracheal extubation was earlier in the tramadol group (P = 0.01). Lengths of pediatric intensive care unit stay did not differ between groups. Times to first trigger of nurse-controlled analgesia bolus and objective pain scores during the 48 h observation period were comparable between groups. The incidence of desaturation and emesis were similar between groups. The patients ate well and did not differ on Day 1 or Day 2.