Antibiotic therapy for the febrile granulocytopenic cancer patient: combination therapy vs. monotherapy.

The specific composition of empiric antibiotic therapy to be implemented for patients with neutropenia and fever remains controversial. The available clinical data comparing antibiotic combination with monotherapy come from studies with a number of major differences in design and analysis that make it difficult to interpret relative efficacy. The standard practice has been to use an antibiotic combination demonstrating broad antibacterial coverage, additive or synergistic effects against the more virulent gram-negative pathogens, and potentially antibacterial activity that decreases the emergence of resistant organisms. Studies of monotherapy have been hindered by wide differences in study design, and few comparative data have been presented to substantiate the efficacy of the drug in question as "treatment of infection"--an exception being the new agent imipenem. The emergence of gram-positive organisms as frequent pathogens in neutropenic patients has raised concern regarding empiric gram-positive therapy. Clinical experience suggests that empiric therapy is unnecessary; rather, directed therapy with an agent such as vancomycin can be used in those situations in which clinical or microbiologic data suggest the need for such antibiotics. If the specifics of optimal management of infectious disease in patients with sustained neutropenia and fever are to be identified, many continuing clinical problems must be approached in a more systematic and defined manner. Defects in study designs of the past should be eliminated, with investigators taking a stronger stand in support of standardized design and analysis.