OBJECTIVE Aplastic anemia is characterized by bone marrow failure and marked reduction of white blood cells, red blood cells and platelets in peripheral blood. Clinical studies have shown that immunosuppressive therapy greatly prolonged the long-term survival of some patients with aplastic anemia. But in severe aplastic anemia (SAA) patients whose ANC was < 0.5 x 10(9)/L, platelets were < 20 x 10(9)/L, very low bone marrow proliferation and high death rate were observed. The present study aimed to evaluate the efficacy of immunosuppressive treatments with cyclosporine A (CSA) alone or CSA combined with antithymocyte globin (ATG) in children with acquired SAA. METHODS Fifty-four cases with SAA were treated with immunosuppressive agents mentioned above in our department from Jan. 1997 to June 2003, 31 of the cases had treated with CSA combined with ATG. There were 18 cases with SAA type I and 13 cases with SAA type II in CSA combined with ATG group, and 13 cases had very severe aplastic anemia. The other 23 cases were treated with CSA alone (CSA group), 10 of these cases had SAA-I and 13 had SAA-II, and 5 cases had very severe aplastic anemia. The responsive rate, relapse, adverse reactions and event free survival (EFS) were compared between CSA combined with ATG group and CSA group. RESULTS The proportions of patients with different types of the disease and severity were comparable between the two groups. The responsive time of the CSA combined with ATG group and CSA group was 2.5 months and 3.5 months, respectively (P < 0.05), the responsive rate in two groups was 81% (25/31) and 52% (12/23), respectively (chi(2) = 4.962, P < 0.05). In 37 cases who were responsive to therapy, the relapse rate was 8% (2/25) and 50% (6/12) respectively (chi(C)(2) = 6.143, P < 0.05). There were no significant differences in adverse reactions to the immunosuppressive agents. All cases were followed-up for more than 1 year, and the event-free survival over one year in these two groups was 81% (25/31) and 52% (12/23), respectively. Forty-seven cases were followed-up for more than two years, and the event-free survival was 74% (20/27) and 50% (10/20), respectively (P < 0.01). Twelve cases were followed-up for over 5 years. There were no secondary tumor, myelodysplastic syndrome and other colony diseases. CONCLUSIONS The immunosuppressive therapies for acquired severe aplastic anemia in childhood were effective. The effect of CSA combined with ATG was better than that of CSA alone, and the relapse rate was lower with the combined treatment. However, the long-term effect needs longer follow-up studies to evaluate.