Empiric treatment of infection during granulocytopenia.

Results from clinical trials conducted over the past 15 years suggest the following: a) Early empiric therapy with broad-spectrum antibiotics directed against Gram-negative bacillary bacteremia is necessary in febrile granulocytopenic cancer patients; b) The level and dynamics of the granulocyte count are extremely important in determining the outcome of bacteremia; c) Most empiric antimicrobial regimens will require therapeutic modifications; these alterations are necessary and contribute to a high overall success rate; d) Only microbiologically documented infections and especially bacteremias are useful for comparison of initial response to antimicrobial regimens; e) The response rate of Gram-negative bacillary bacteremia is clearly influenced by the susceptibility of the causative pathogen to the beta-lactam component of the empiric regimen; emergence of resistance to some beta-lactam antibiotics is quite common and necessitates successive modifications of empiric regimens with time; f) The combination of an anti-pseudomonal beta-lactam with an aminoglycoside is recommended as the standard for empiric therapy in febrile granulocytopenic cancer patients, especially in those with severe and persistent granulocytopenia who are suspected of having Gram-negative bacillary bacteremia; less neutropenic and/or asymptomatic patients may do well with monotherapy; g) Gram-positive pathogens have become a common cause of bacteremia in granulocytopenic cancer patients; the response rate to empiric regimens may be suboptimal but the associated mortality is low; h) Patients with severe granulocytopenia and protracted fever whose blood cultures remain negative are at high risk for contracting fungal infections; in these patients, empiric antifungal agents are probably indicated.