Objective To investigate the clinical efficacy and prognostic factors of double autologous hematopoietic stem cell transplantation (DAHSCT) in treatment of aggressive non-Hodgkin lymphoma (NHL). Methods From March 2001 to December 2014, a total of 48 patients with aggressive NHL who underwent scheduled DAHSCT in Tangdu Hospital, the Fourth Military Medical University were selected as the study subjects. Among them, 29 patients were male and 19 cases were female; and 22 patients were with B-cell non-Hodgkin lymphoma (B-NHL), 26 cases were with T-cell non-Hodgkin lymphoma (T-NHL). All 48 patients underwent the second autologous hematopoietic stem cell transplantation (auto-HSCT) within 2 to 6 months after receiving the first auto-HSCT. Retrospective analysis was used to collect clinical data from all patients, including age, occurrence of B symptoms, clinical staging of Ann Arbor/Cotswolds staging system, level of serum lactate dehydrogenase (LDH), international prognostic index (IPI) score, occurrence of extra-nodal lesions, and remission status of disease before the first and second auto-HSCT, etc.. Kaplan-Meier method was used to analyze the survival of 48 NHL patients. Univariate Cox regression analysis was used to analyze the factors that may influence the overall survival (OS) period of NHL patients after DAHSCT. According to the available research results and clinical experience, the further multivariate Cox regression analysis was carried out for the factors that were statistically significant in the univariate analysis, as well as some factors that were not statistically significant in the univariate analysis, but with clinical significance. Results ① All 48 aggressive NHL patients were collected enough peripheral blood stem cells, and got hematopoietic reconstitution after receiving auto-HSCT. In this study, the median number of CD34+ cells transfused was 3.40×106/kg [(2.70-8.70)×106/kg] in the first auto-HSCT and 3.68×106/kg [(2.95-8.00)×106/kg] in the second auto-HSCT, respectively. And the median time of platelet reconstruction was 12.0 d (9.0-17.0 d) in the first auto-HSCT and 14.0 d (9.0-18.0 d) in the second auto-HSCT, respectively. The median time of neutrophil reconstruction after transplantation was 12.0 d (9.0-16.0 d) in the first auto-HSCT and 13.0 d (9.0-16.0 d) in the second auto-HSCT, respectively. ② For all 48 patients, the overall response rate of DAHSCT was 93.8% (45/48) after receiving DAHSCT. The median follow-up time was 78.0 months (42.0-105.0 months). In this study, the rates of disease-free survival, recurrence and mortality were 64.6% (31/48), 16.7% (8/48) and 22.9% (11/48), respectively. ③ In this study, 5-year OS rate was 76.6% and 5-year progression-free survival (PFS) rate was 65.8%. ④ Results of univariate Cox regression analysis showed that achievement of complete remission (CR) before the second auto-HSCT was a protective influencing factor of OS period for aggressive NHL patients (HR=0.174, 95%CI: 0.051-0.598, P=0.005). And multivariate Cox regression analysis of factors that may influence the OS period of NHL patients showed that achievement of CR before the second auto-HSCT was an independent protective influencing factor of OS period for aggressive NHL patients (HR=5.046, 95%CI: 1.031-24.683, P=0.046). Conclusions DAHSCT treatment for aggressive NHL has good efficacy, especially for the patients with CR before the second auto-HSCT. However, whether DAHSCT deserves further clinical application in treatment of aggressive NHL patients still needs further study to confirm. Key words: Lymphoma, non-Hodgkin; Hematopoietic stem cell transplantation; Transplantation, autologous; Autologous hematopoietic stem cell transplantation