An open-label, randomized comparative study was conducted to evaluate the efficacy and safety of cefepime (2.0 g q. 8 h) and ceftazidime (2.0 g q. 8 h) in the empiric therapy of febrile neutropenic patients. A total of 45 eligible febrile episodes were randomized (1:1) to be treated with the study regimen. Nineteen febrile episodes treated with cefepime and 22 febrile episodes treated with ceftazidime were evaluable for efficacy. The two groups were comparable in terms of age, sex, height, weight, underlying neoplasm, number of pretherapy neutrophil, duration of neutropenia and types of infections. The overall therapeutic success rate of the cefepime group (53%) was comparable to the ceftazidime group (50%). It did not differ significantly (95% confidence interval: -0.28 to 0. 34, p = 0.85). Eighty-eight percent of pathogens in each group were bacteriologically eradicated. The safety profile was similar in both groups. No patients in either group discontinued the therapy because of adverse events. None (0%) of the cefepime patients and 2 (9%) of the ceftazidime patients died of infection. The results of this study suggest that cefepime is an effective and safe agent in the empiric therapy of febrile episodes in neutropenic patients.