Febrile neutropenia is one of the major causes of morbidity, treatment interruptions and mortality during oncological treatment and bone marrow transplant (BMT). The time from fever to initiation of empiric antibiotics therapy (AET) is related to the outcome of patients. Ideally, patients should receive EAT in less than one hour after fever. However, many patients fail to receive EAT in less than 1 hour due to intrahospitalar delays, and may have impact in patient outcomes. Therefore, institutional policies to ensure ideal management of these patients are recommended and fever-to-patient antibiotic delivery is one of quality control measure in our Oncology and BMT Unit.


We implemented a nurse-based program to reduce the time from first fever or clinical EAT indication to EAT infusion. A senior nurse was in charge of implementing the program in the Oncohematological and Bone Marrow Transplantation Unit. Several actions were implemented: (1) Data collection for comparative studies, (2) Team education (pharmacy, medical and nurses) through lectures and (3) Daily data verification to verify the accuracy of data registered in medical files; (4) Survey about patient characteristics included in this study.

All patients in the unit were daily censored for neutropenia, and neutropenic patients were followed for fever or need for ATB initiation. Patients who initiated EAT on the emergency room (ER) or day clinic were excluded from the analysis. For all patients, time from EAT indication and EAT infusion was collected. Data was then classified in 4 categories: (1) Time to EAT <1 hour; (2) Time to EAT >1 hour; (3) Major data inconsistency (defined when EAT indication registered time was later than EAT infusion registered time); (4) Minor data inconsistency (defined when time to EAT infusion was registered at the same time of EAT indication). Twenty medical files were retrospectively selected for comparison.


We present the results of the first three months after the program implementation. In the retrospective group, only 35% (7/20) of patients received EAT in less than 1 hour, 15% (3/20) of patients received EAT in more than 1 hour, and inconsistencies were seen in 50% of medical files, including 40% of minor and 10% major inconsistencies. After program implementation (n=17), the percentage of patients receiving EAT in less than 1 hour was 82% (14/17), 6% of patients received EAT >1 hour (1/17), and 12% (2/17) of patients had minor inconsistencies in registered time. No major inconsistencies were observed after the program.

For good practicing analysis, we grouped patients in two groups: Ideal (EAT <1 hour) and Not Ideal (EAT>1 hour and inconsistencies in medical file registration). With this approach, we showed that Ideal group improved from 35% to 82%, and the Not Ideal group declined from 65% to 18% (p=0.007). Moreover, the median time to EAT decreased from 60 minutes (range 30-240) to 30 minutes (range 8-66) (p=0.01)


The implementation of a nurse-based program significantly increased the number of patients receiving ATB in <1 hour in an Oncohematological and Bone Marrow Transplant Unit after only 3 months. Moreover, the time to ATB initiation was significantly decreased with this policy. Our findings indicate that, despite the nursing staff recognize the importance of febrile neutropenia, monitoring process, education and constant communication are necessary for an effective treatment and for improvement the patient care. Further implementation of this program in the day clinic and ER are planned and a survey about patient education of neutropenic infections will be implemented.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.