OBJECTIVE To evaluate the efficacy and safety of prophylactic ceftazidime on early bacterial infection in APBSCT recipients during neutropenia. METHODS APBSCT recipients were prospectively randomly assigned to intravenous ceftazidime treatment group and control group (no prophylaxis of antibiotics). The treatment started from the first day until resolution of neutropenia or the appearance of early bacterial infection. RESULTS From March 2010 to January 2013, 70 APBSCT recipients were enrolled in the study with 36 in treatment and 34 in control group. Overall, 29(41.4%) patients developed early bacterial infection, among which, 9(25.0%) in the treatment group and 20(58.8%) in the control group (P = 0.004). The median infection free survival (IFS) was not reached in the treatment group and was 8 days in the control group (P = 0.005). Despite whether patients received single high dose melphalan or other conditioning regimes, the early bacterial infection rate was lower in the treatment group than in the control group, and the median IFS was longer in the treatment group than that in the control group. The mean courses of antibiotic administration were (8.08 ± 2.03) days and (3.68 ± 3.56) days respectively in the treatment and control groups (P < 0.001). However, the duration of empirical carbapenems were (1.67 ± 3.03) days and (3.68 ± 3.56) days respectively (P = 0.013). There was no significant difference of antibiotics cost per patient between the two groups. Four patients in the treatment group had a transient elevated serum creatinine. Overall, no infection related mortality was observed in either group. CONCLUSIONS Prophylaxis of intravenous ceftazidime for APBSCT recipients is effective in preventing early bacterial infection with an acceptable toxicity and cost profile. However, it doesn't have effect on infection related mortality. Therefore, our results do not support the use of antibiotic prophylaxis for patients undergoing APBSCT.