JOURNAL OF HEMATOTHERAPY & STEM CELL RESEARCH 9:645–649 (2000) Mary Ann Liebert, Inc. State-of-the-Art Review Clinical and Laboratory Features Predicting a Favorable Outcome and Allowing Early Discharge in Cancer Patients with Low-Risk Febrile Neutropenia: A Literature Review ELIO CASTAGNOLA, 2 DIMITRI PAOLA, 1 RAFFAELLA GIACCHINO, 2 and CLAUDIO VISCOLI 1 ABSTRACT To value feasibility of early discharge in febrile granulocytopenic patients, 27 original paper pub- lished in the last 11 years were analyzed concerning these clinical and therapeutic approaches. A Medline search of English language literature published in the last 11 years (1988–1999) used the key words neutropenia, fever, cancer, home-antibiotic therapy, short course of antibiotic therapy, and early discharge. Twenty-seven original papers fulfilling the study criteria were identified. In these studies, 5208 episodes were evaluated: there were 538 failures with 87 deaths. Features of low- risk patients who developed life-threatening infectious disease were related to general clinical con- dition, cancer control, bone marrow function, presence of clinical signs of infection, and social fea- tures. Careful risk assessment can allow safe recognition of low-risk patients with febrile neutropenia who can be discharged early and can be used to follow outpatient treatment programs to improve patients’ quality of life as well as the use of economic resources. 645 1 Unit for the Infections in the Immunocompromised Host, National Cancer Institute, University of Genova, Genova, Italy. 2 Infectious Diseases Unit, G. Gaslini Children’s Hospital, Genova, Italy. INTRODUCTION M ANAGEMENT OF FEVER in neutropenic cancer pa- tients remains one of the most frequent and po- tentially life-threatening problems in everyday clinical oncology. In 1966, Bodey et al. (1) showed that absolute neutrophil count (ANC) is strictly related to the risk and severity of infectious complications in this patient popu- lation. Many of the clinical reports published in the past 25 years clarified that nearly two-thirds of febrile episodes in neutropenic cancer patients have an infec- tious etiology. In these patients fever is commonly the first and often the only sign of infection, since other clin- ical findings can be lacking because of the atypicalness of inflammatory response due to neutropenia. For many years, the standard management of febrile neutropenic patients involved hospitalization of the pa- tients so that they could receive empirical broad-spec- trum parenteral antibacterial therapy. Although this treat- ment strategy has evolved from experience in high-risk patients with hematological malignancies, in whom bac- terial infection can result in substantial morbidity and early mortality, it has been adopted for all patients with febrile neutropenia, largely because of the inability of clinicians to distinguish reliably and rapidly between pa- tients who are at high risk for developing life-threaten- ing bacterial infections and those with fever of unknown origin (FUO). Recently, the development of risk-assessment models has facilitated the recognition of high-, moderate-, and low-risk subgroups among febrile neutropenic patients and allows early discharge with or sometimes without outpatient antibiotic therapy for low-risk groups, with the same degree of efficacy and safety as hospital-based ther- apy and the best quality of life for patients and expense allocation of public health administration. However, knowledge about risk factors predicting favorable out- comes of febrile neutropenia in cancer patients is still