nerve block failed, we gave routine medication for pain. The time when the patient first complained of pain was recorded. The anesthesiologist visited the children and parents at 8 and 16 hours after the operation to evaluate visual analogue scale (VAS 0-10, > 8 years) or faces pain scale (FPS, < 8 years). We also reviewed medical records for routine analgesics used during the hospital stay, and incidence of nausea or vomiting. Brain MRI was performed if any signs or symptoms of brain infarction were observed. Patients who recovered from the operation without complications made follow-up visits after 6 months, at which brain MRI scans were performed. Sample size was determined according to a pilot study of 13 patients performed in 2006. The time at which patients first complained of pain was 12.9 ± 5.3 hours in group T and 6.3 ± 5.2 hours in group R. If we hypothesize that a difference in block times between groups exists, we would need at least 9 subjects in each group at a significance level of 5% and power of 80%. We used SPSS 12.0 for statistics analysis; the unpaired T-test was used to compare continuous variables between two groups. Pearson’s Chi square was used to determine difference of frequency of nausea, vomiting or brain infarction after the operation. P < 0.05 was considered statistically significant. There was no difference in demographic characteristics, preoperative infarction history, operation time, anesthesia time, PACU stay time, or medication during the postoperative period. Group R patients first complained of pain at 9.2 ± 9.4 hours and Group T patients at 14.2 ± 10.0 hours (P = 0.11). VAS at 8 hours after surgery was 1.2 ± 1.5 in group R and 1.5 ± 2.4 in group T (P = 0.56); at 16 hours it was 1.2 ± 1.7 in group R and 2.3 ± 2.8 in