Pediatric patients with acute myeloid leukemia (AML) mainly receive myeloablative conditioning regimens based on busulfan (BU) or total body irradiation (TBI) before allogeneic hematopoietic cell transplantation (allo-HCT); however, the optimal conditioning regimen remains unclear. To identify which of these regimens is better for pediatric patients, we performed a retrospective analysis of nationwide registration data collected in Japan between 2006 and 2011 to assess the outcomes of patients receiving these regimens before a first allo-HCT. Myeloablative conditioning regimens based on i.v. BU (i.v. BU-MAC) (n = 69) or TBI (TBI-MAC) (n = 151) were compared in pediatric AML patients in first or second complete remission (CR1/CR2). The incidences of sinusoid obstruction syndrome, acute and chronic graft-versus-host disease, and early nonrelapse mortality (NRM) before day 100 were similar for both conditioning groups; however, the incidence of bacterial infection during the acute period was higher in the TBI-MAC group (P = .008). Both groups showed a similar incidence of NRM, and there was no significant difference in the incidence of relapse between the groups. Univariate and multivariate analyses revealed no significant differences in the 2-year relapse-free survival rates for the i.v. BU-MAC and TBI-MAC groups in the CR1/CR2 setting (71% versus 67%, P = .36; hazard ratio, .73; 95% CI, .43 to 1.24, respectively). TBI-MAC was no better than i.v. BU-MAC for pediatric AML patients in remission. Although this retrospective registry-based analysis has several limitations, i.v. BU-MAC warrants further evaluation in a prospective trial.