Background: HSCT from alternative donor (AD-HSCT) is indicated as salvage therapy for pediatric patients with severe aplastic anemia (SAA) who failed immunosuppressive therapy (IST) and lack a MSD. Given with the recent advances in AD-HSCT outcomes, the indication of AD-HSCT in SAA needs to be reevaluated. Methods: We retrospectively reviewed the medical records of 45 pediatric patients with SAA who received HSCT at Asan Medical Center Children’s Hospital from February 2001 to January 2014. To analyze the efficacy of HSCT from alternative donors, we compared overall survival, incidence of acute and chronic GVHD, incidence of graft failure (GF), and incidence of post HSCT infections including CMV reactivation, PTLD, bacteria and fungi according to the type of donors. Results: Of the 45 patients, 32 were male and 13 were female. Median age at diagnosis was 8.0 years (range, 0.2-17.0). Median age at HSCT was 12.9 years (range, 0.7-21.7). Median CD34+ cell dose was 7.64 X 106 /kg (range, 2-33.8). Ten patients received HSCT from MSD (MSD-HSCT), 15 patients from URD (URD-HSCT) and 20 patients from haploidentical family donor (HFD-HSCT). Twenty-six patients (55.3%) had received IST prior to HSCT. Of 45 patients, eight patients experienced GF. Cumulative incidence (CI) of GF was higher in HFD-HSCT compared to the other type of HSCT (30% vs 8%, P=0.049). Among the 8 patients, 6 patients achieved engraftment after additional HSCT, one died of infection after GF and the remaining one is waiting for the second HSCT. The CIs of acute GVHD were 20% for MSD-HSCT, 60% for URD-HSCT and 40% for HFD-HSCT, respectively ( P＞0.05). Twenty-two patients developed CMV reactivation, 7 had PTLD and 13 had bacterial infection after HSCT. CI of CMV reactivation was higher in AD-HSCT than that of in MSD (63% vs 0%, P=0.002). Three patients died; one died of acute GVHD, one died of autoimmune hemolytic anemia and the remaining one died of infection after GF. At a median follow-up of 40.8 months (range, 6.5-162.8), 3 year-OS was 92.9% (100% for MSD-HSCT, 92.3% URD-HSCT and 89.4% for HFD-HSCT, P＞0.05). All surviving 42 patients survived were transfusion-independent. Conclusions: Our study suggest that early alternative HSCT including HFD-HSCT could be beneficial to the children with SAA who do not have MSD. However, a larger multicenter study is needed to verify our results..