Early discontinuation of empirical antibacterial therapy in febrile neutropenia: the ANTIBIOSTOP study

Abstract

Introduction: Immediate empirical antibiotic therapy is mandatory in febrile chemotherapy-induced neutropenia, but its optimal duration is unclear, especially in patients with fever of unknown origin (FUO).

Objectives: The primary objective of this 20-month prospective observational study was to evaluate the feasibility and safety of short-term antibiotic treatment in afebrile or febrile patients exhibiting FUO, irrespective of their neutrophil count. The secondary objective was to describe the epidemiology of all episodes of febrile neutropenia.

Methods: In the first phase of the study, empirical antibiotic therapy in FUO patients was stopped after 48 h of apyrexia, in accordance with European Conference on Infections in Leukaemia guidelines (n = 45). In the second phase of the study, antibiotics were stopped no later than day 5 for all FUO patients, regardless of body temperature or leukocyte count (n = 37).

Results: Two hundred and thirty-eight cases of febrile neutropenia in 123 patients were included. Neither the composite endpoint (p = .11), nor each component (in-hospital mortality (p = .80), intensive care unit admission (p = 0.48), relapse of infection ≤48 h after discontinuation of antibiotics (p = .82)) differed between the two FUO groups. Violation of protocol occurred in 17/82 episodes of FUO without any major impact on statistical results. Twenty-six (57.3%) and 22 (59.5%) FUO episodes did not relapse during hospital-stay (p = 1), and nine (20%) and five (13.5%) presented another FUO, respectively. One hundred and fifty-six episodes of febrile neutropenia (65.5%) were clinically or microbiologically documented, including 85 bacteremia.

Conclusions: These results suggest that early discontinuation of empirical antibiotics in FUO is safe for afebrile neutropenic patients.

Keywords:

• Febrile neutropenia
• Antibiotic therapy
• Bacterial epidemiology
• Malignant haemopathy

Acknowledgements

The authors would like to thank Emmanuel Nowak for his statistical work (INSERM CIC-1412, Brest Teaching Hospital, France) and all paramedical staff for their assistance in the management of haematological patients.

Disclosure statement

The authors declare that they have no conflict of interest.