Antithymocyte globulin (ATG) as part of conditioning regimens is known to reduce the incidence and severity of acute and chronic graft-versus-host disease (aGVHD, cGVHD). The influence of ATG on transplant-related mortality (TRM) and disease-free survival (DFS) is controversial, and may depend on the dose and timing of ATG. We retrospectively compared 2 doses of ATG-Fresenius (ATG-F) in patients undergoing matched unrelated donor allogeneic hematopoetic stem cell transplantation (HSCT) for hematologic malignancies. A dose of 60 mg/kg body weight has previously been recommended for ATG-F. All patients received cyclosporine A and short course methotrexate. ATG-F was administered at a dose of 30 mg/kg on day -1 (ATG-30 group, n = 34) or 20 mg/kg/day on days -3 to -1 (ATG-60 group, n = 49). There was no difference in time to leukocyte and platelet engraftment in the 2 groups. The incidence of aGVHD grade II-IV (50% versus 53%, P = .83) and grade III-IV (27 versus 20%, P = .60) was similar in the ATG-30 versus ATG-60 groups, respectively. There was a trend to a higher incidence of cGVHD in the ATG-30 group (59% versus 40%, P = .14). The estimated 3-year incidence of relapse was similar in the ATG-30 and ATG-60 groups (15% versus 16%, P = .84) whereas the 2-year TRM was lower for the ATG-30 group (12% versus 33%, P = 0.02), mainly because of a higher incidence of fatal infections in the ATG-60 group. This resulted in a better DFS (73% versus 51%, P = .07) for the ATG-30 group. ATG-F (30 mg/kg) administered as a single dose on day -1 may lead to better outcome in patients undergoing unrelated donor allogeneic HSCT compared to 60 mg/kg given in 3 equivalent doses. A prospective randomized study comparing these 2 doses of ATG-F is warranted.