Background: Infection remains a major cause of morbidity and mortality in febrile neutropenic patients. The empiric use of cefepime plus amikacin could provide a synergistic effect against bacteria. We have previously demonstrated that this antibiotic regimen was successful in reducing antibiotic resistance rates, that emerged under a previous regimen of ceftazidime + glycopeptide as empiric start therapy for neutropenic fever (J. Mebis et al. Leukemia 1998; 12:1627-29). Methods: We performed a prospective single-centre study of hematological patients with febrile neutropenia, treated with cefepime (3 x 2 g/d IV) and amikacin (15 mg/kg/d IV). In case of persistent fever after 48 hours, a glycopeptide (vancomycin or teicoplanin) was added. After an additional 48 hours cefepime was replaced by a carbapenem (imipenem or meropenem). Amphotericin B was initiated if fever persisted. The antibiotic regimen was considered to be successful in case of resolution of the fever before initiation of a carbapenem became necessary, without recurrence of fever and with resolution of all clinical and microbiological signs under the existing regimen (cefepime + amikacin +/- glycopeptide). Results: A total of 200 febrile episodes in 119 neutropenic hematological patients have been analyzed. In 68 episodes (34%) the etiology of the fever was microbiologically documented: 57 bacteremias (28.5%) - with a majority of gram-positive pathogens - with in 5 cases concurrent pneumonia and in 11 episodes (5.5%) pathogens have been isolated from other sources than hemocultures. A clinically documented infection was demonstrated in 56 episodes (28%): predominantly pneumonia (n = 19) and oral mucositis (n = 19). Seventy-six febrile episodes (38%) could not be documented clinically or microbiologically. Eighty-six episodes (43%) resulted in defervescence after initiating cefepime plus amikacin and a further 36 (18%) after addition of a glycopeptide. In 12 episodes (6%) the patient became afebrile after replacing cefepime by a carbapenem and in 4 episodes (2%) after subsequent addition of amphotericine B. In 58 episodes (29%) other antimicrobial drugs had to be started to obtain defervescence because of recurrent fever after a short afebrile period (n = 23) or because of persisting fever (n = 35). Two patients (1%) died because of a severe infection. Two febrile episodes (1%) were not evaluable. No major antibiotic-related side effects were noted. Conclusion: The empiric use of the antibiotic regimen consisting of cefepime + amikacin in febrile neutropenic patients led to resolution of the fever in 43% of the studied cases, increasing to 61% after the addition of a glycopeptide. Our results are broadly concordant with data from international literature, demonstrating the efficacy of this empiric antibiotic regimen in high-risk neutropenic patients.
Full conference title:
American Society of Hematology 45th Annual Meeting
- ASH 45th (2003)