We undertook a randomized double-blind trial to compare the efficacy of 1.5 mg/kg body weight (low dose) and 3 mg/kg (moderate dose) lidocaine regional anesthesia for closed reductions of forearm fractures in childhood. Of the 283 children studied, 143 were randomized to the moderate-dose group and 140 to the low-dose group. The characteristics of the children and their injuries did not differ significantly. There were no complications due to lidocaine toxicity. In children with angulated and incompletely displaced fractures, satisfactory anesthesia was achieved in 94% of those receiving the low dose and in 97% of those receiving the moderate dose of lidocaine. In children with completely displaced fractures, satisfactory anesthesia was achieved in 93% of those receiving the moderate dose but in only 67% of those receiving the low dose of lidocaine. We conclude that the low-dose lidocaine protocol is suitable for children requiring closed reductions of angulated and incompletely displaced fractures of the forearm. In contrast, the moderate-dose lidocaine protocol is more reliable in children with displaced forearm fractures. Meticulous adherence to the protocols is essential to prevent systemic lidocaine toxicity from premature deflation of the tourniquet. This potential risk is further reduced by use of the low-dose protocol, which is applicable to approximately 70% of the children with forearm fractures requiring closed reductions.