2688. The Clinical Impact of Early De-escalation of Broad-Spectrum Antibiotics in Acute Myeloid Leukemia Patients with Febrile Neutropenia

BACKGROUND: In patients with febrile neutropenia (FN) the initiation of broad-spectrum antibiotics (BSA), an anti-pseudomonal agent +/− vancomycin, is recommended by national guidelines. BSA should be continued until absolute neutrophil count (ANC) recovery (ANC > 500 cells/mm(3)). With increasing antimicrobial resistance, clinicians are reassessing the need to continue BSA until count recovery; new data are emerging that patients may be able to have their BSA de-escalated if stable and afebrile. At our institution, some patients are de-escalated from BSA to a fluoroquinolone before ANC recovery and others are continued on BSA. The purpose of this study was to evaluate the efficacy and safety of early de-escalation compared with the standard of care. METHODS: We retrospectively reviewed acute myeloid leukemia patients receiving induction chemotherapy who developed FN while at Yale New Haven Hospital from March 2013 to August 2018. Patients were excluded if they developed a culture documented infection, received incomplete or multiple induction chemotherapy treatments, or died from underlying disease during hospitalization. The primary outcome was recurrent fever during admission and secondary outcomes included incidence of breakthrough infections (BI), duration of hospital stay, early discharge (discharge before ANC recovery), duration of BSA, and readmission within 7 days of discharge. RESULTS: A total of 210 patients were evaluated and 91 patients were included (de-escalation, n = 45; BSA, n = 46). Baseline characteristics are noted in Table 1. There was no statistical difference in rate of recurrent fever in patients who were de-escalated from BSA compared with those that were continued (P = 0.05). De-escalated patients had a shorter duration of BSA therapy (P < 0.05), earlier discharge (P = 0.05) and no difference in readmission rates (P = 0.39) (Table 2). There was no difference in rate of BI between both groups and all BI were bacteremias. (Table 3) No patients who experienced a BI died from infection. CONCLUSION: The results of this study revealed no difference in the primary outcome of recurrent fever between the BSA and de-escalation groups. De-escalation led to a reduced duration of BSA and facilitated earlier discharge without increasing readmission rates and BI. [Image: see text] [Image: see text] [Image: see text] DISCLOSURES: All authors: No reported disclosures.