For young patients, HLA-MRD HSCT is the first-line treatment of SAA. However, due to China's birth control policy, few patients could find suitable sibling donors and HLA-MUD. More and more transplantation centers have used Haplo-D as the donor source for young adult and pediatric patients. However, studies with larger amount of pediatric patients are rare. We retrospectively analyzed the data of children with AA who were treated with allogeneic HSCT and compared the therapeutic efficacy of Haplo-HSCT and MRD/MUD group. A total of 62 patients were enrolled. Implantation was successfully performed in 58 patients. There was no significant difference in the time for reconstruction of hematopoietic function between patients in the two groups. Thirty-two had grade I-IV aGVHD with incidence of 51.61%. The incidence of aGVHD was 79.41% for patients in the Haplo-HSCT, significantly higher than that of 17.86% for patients in the MRD/MUD group (P < .01). However, the incidence of cGVHD was not significantly different between patients in the two groups (26.47% vs 10.71%, P = .09), the incidence of CMV infection was 28.57% and 52.94% for patients in the MRD/MUD and Haplo group, respectively, showing no significant difference (P = .053). The incidence of EBV infection was 47.06% for patients in the Haplo group and 28.57% for patients in the MRD/MUD group, showing no significant difference (P = .11). However, the 3- and 5-year cumulative OS and FFS rates showed statistically significant difference in the two groups, P = .012 and .045, respectively. Compared to Haplo-HSCT, MRD/MUD is more economic. In this study, we achieved good Haplo transplantation results. The incidences of cGVHD and CMV/EBV were not significantly different between Haplo group and MRD/MUD group. Although OS and FFS of the Haplo group were not as good as those of the MRD/MUD group, it is still acceptable as an alternative treatment under emergency.