Injured spleens have been successfully managed without operation in a number of children; however, splenectomy or splenic-conserving surgery may not be avoided because of exsanguinating hemorrhage. This study was performed to evaluate the efficacy of splenic arterial embolization (SAE) to control hemorrhage from injured spleens in children. We compared the outcomes of two groups of children with splenic injury. The first group (G1) consisted of eighteen children who were managed with conventional selective nonoperative treatment between 1993 and 1994. The second group (G2) consisted of 23 children prospectively studied from 1996 to 1997 after SAE was added in the management protocol of splenic injury. The criteria for SAE were grade III or IV injury, extravasation of contrast material revealed by CT, or unstable vital signs without evidence of associated injuries. Laparotomy was performed in 6 patients of G1 (33.3%), 2 of whom had associated injuries. Five underwent splenectomy and the overall salvage rate in G1 was 72.2% (13/18). In G2, eight patients (34.8%) had SAE, which stopped bleeding successfully in all patients. Two of G2 (8.7%) had laparotomy because of associated injuries. Only one patient underwent splenectomy and the salvage rate was 95.6% (22/23). No patients required transfusion after SAE. In conclusion, the SAE effectively controlled hemorrhage from injured spleens. More spleens were salvaged with a reduced laparotomy rate after application of SAE in splenic injury.