Empiric therapy of bacterial infections in severe neutropenia

Abstract

Infection remains the principal cause of morbidity and mortality in febrile neutropenic episodes in cancer patients. During the 1960s and 1970s, infection was responsible for 80% of deaths in patients with hematological malignancy, but observations from clinical trials completed by the International Antimicrobial Therapy Cooperative Group of the European Organisation for Research and Treatment of Cancer (IATCG-EORTC) show that in recent years, mortality rates due to infection have fallen to approximately 3% in patients undergoing active treatment of the malignancy. For patients considered beyond active treatment, figures are much higher. This improvement is undoubtedly due to the introduction and use of empirical therapy in the early 1970s. The rationale for developing the concept of empirical therapy was based on the fact that blood culture results may not be available for several days, with disastrous consequences for treatment of a rapidly progressing infection. Therefore, antibiotics are given to febrile neutropenic cancer patients without definitive microbiological proof of infection. Many studies on empirical therapy have investigated combinations of agents, such as a β-lactam plus an aminoglycoside. Such antibiotic combinations offer broad-spectrum antibacterial coverage and have been shown to provide synergistic bactericidal effects against Gram-negative pathogens in vitro. However, in recent trials on the comparison between mono- and combination therapy the time to defervescence did not significantly differ, suggesting that a more synergistic bactericidal effect, as observed in vitro with combinations, did not have a measurable counterpart in vivo on fever duration. Combinations may also reduce the emergence of resistant isolates that are observed when certain β-lactam antibiotics are used alone. Aminoglycoside-containing regimens are, however, associated with ototoxicity and nephrotoxicity. In efforts to circumvent this, some investigators have used double β-lactam combinations but, recently, monotherapy with newer extended-spectrum agents such as the carbapenems, third- and fourth-generation cephalosporins and fluoroquinolones have become increasingly studied and used as standard empirical therapy. The criteria for evaluating the response to empirical antibiotic treatment in febrile neutropenia have evolved from a number of clinical trials. These studies have indicated that age, the duration of neutropenia, the nature of the underlying disease, the causative pathogen, the availability of new effective drugs, and the chosen end point all have a profound influence on response rates. Moreover, a clear understanding of the objectives and design of trials is of paramount importance in interpretation of the results and comparison of data from different trials. Recently, there has been a move towards more rational use of antibiotics with the identification of patient subgroups at lower risk of unfavourable outcome, who might be treated more appropriately with less aggressive, tailored, empirical therapy. This review examines the effectiveness of empirical therapy in studies from different centers during the last 20 years, and looks towards possibilities for modification of treatment strategies in the future, to combat emergence of new pathogens and the development of resistance in established pathogens.