Is current initial empirical antibiotherapy appropriate to treat bloodstream infections in short-duration chemo-induced febrile neutropenia?

Fever of unknown origin is by far the most common diagnosis in low-risk febrile neutropenic patients undergoing chemotherapy. The current empirical regimen combines amoxicillin-clavulanic acid and fluoroquinolones in low-risk neutropenic patients. The aim of this study was to assess the appropriateness of antibiotherapy and the outcome of bloodstream infections (BSI) in patients with expected neutropenia of short duration. This 2-year monocentric retrospective study included all consecutive neutropenic febrile adult patients with expected duration of neutropenia ≤ 7 days. They were classified into low- and high-risk groups for complications using the MASCC index. Appropriateness of initial empirical antibiotic regimen was assessed for each BSI. Multivariate analysis was performed to identify factors associated with mortality. Over the study period, 189 febrile episodes with positive blood cultures in neutropenic patients were reported, of which 44 occurred during expected duration of neutropenia ≤ 7 days. Patients were classified as high-risk (n = 27) and low-risk (n = 17). Gram-negative bacteria BSI represented 57% of cases, including only two multidrug-resistant bacteria in high-risk patients. Initial empirical antibiotherapy was appropriate in 86% of cases, and inappropriate in the event of coagulase-negative Staphylococcus BSI (14%), although the outcome was always favorable. In low-risk patients, no deaths and only 12% of severe complications were reported, contrasting with mortality and complication rates of 48% (p < 0.001) and 63% in high-risk patients (p < 0.001), respectively. Outcome of BSI is favorable in low-risk febrile neutropenic patients, even with inappropriate empirical initial antibiotic regimen for coagulase-negative Staphylococcus BSI. Initial in-hospital assessment and close monitoring of these patients are however mandatory.