Background: Infectious complications are an important cause of morbidity and mortality in cancer patients, especially those receiving chemotherapy. Early empirical administration of broad-spectrum antibiotics at the onset of fever has become common practice, but the specific empirical regimen remains controversial. Over the last two decades, a marked shift in the spectrum of causative organisms towards a gram-positive predominance has been the main factor influencing therapeutic approaches.
Methods: A MEDLINE search (January 1966-March 1999) for studies evaluating the treatment of adult patients with febrile neutropenia was carried out. Reference lists from identified articles also served as literature sources. All studies were critically evaluated for an evidence-based discussion.
Results: Most clinical studies have advocated the use of combination therapy, including a beta-lactam plus an aminoglycoside. However, recent trials comparing combination therapy versus monotherapy with the newer extended-spectrum agents such as carbapenems and third-generation cephalosporins could not prove relevant differences in outcome. Empirical therapy has evolved towards a planned-progressive modification in patients with persistent fever, especially aiming at early antifungal coverage. Although overwhelming streptococcal sepsis has raised particular concern, the role of glycopeptide antibiotics remains another controversial issue especially because of increasing reports of infections caused by resistant grampositive cocci (enterococci and recently S. aureus). Likewise routine antibacterial and antifungal chemoprophylaxis is being questioned due to its association with emergence of resistant organisms (particularly fluoroquinolone-resistant Escherichia coli and fluconazole-resistant non-albicans Candida species).
Conclusions: Empirical therapy has dramatically reduced mortality rates in febrile neutropenic patients. Monotherapy with extended-spectrum antibiotics is a feasible alternative to combination therapy. The indiscriminate use of empirical glycopeptides should be discouraged. Early antifungal therapy with amphotericin B is essential in case of persistent fever. Prompt recognition of institutional resistance trends is of paramount importance for an optimal antibiotic selection. Further studies of larger size designed to confirm low-risk patient characteristics are necessary.