A prospective, randomized, multicenter study was conducted to compare the efficacies of cefepime and ceftazidime in combination with amikacin as initial empirical antibiotic regimen for the treatment of febrile neutropenic patients with hematologic malignancies. A total of 118 patients with acute leukemia or non-Hodgkin lymphoma were randomized to receive either cefepime (2 g b.i.d.) or ceftazidime (2 g t.i.d.) in combination with amikacin (1 g/day). At the end of the study, 55 patients in each group were evaluable for efficacy. The groups consisted of patients with similar demographic characteristics and underlying illnesses. In 34 patients (31 %), the infection could be clinically or microbiologically documented. The median duration of antibiotic therapy was 8 days and the median time of defervescence of fever in responders was 2 days for both groups. 15 (27.3 %) of the patients in the ceftazidime group and 20 (36.3 %) in the cefepime group responded to the empirical regimen. Modification of the initial regimen with antivirals and/or azole antifungals raised the number of responders to 19 (34.5 %) and 27 (49.1 %), respectively. However, the difference between the success rates of the two groups did not reach statistical significance. Three patients in the ceftazidime group and 5 in the cefepime group experienced mild and reversible study drug-related side effects, such as skin rash and nephrotoxicity due to amikacin use. In conclusion, cefepime in combination with amikacin is at least as effective as ceftazidime and amikacin in this setting.