Early Carbapenem De-escalation for Patients With High-Risk Febrile Neutropenia and Hematologic Malignancy

Background: Antibiotic stewardship programs recommend an early antibiotic de-escalation strategy in patients with hematologic malignancyassociated febrile neutropenia without documented infectious foci. Studies
on its safety are scarce so there is low adherence to the recommendation.

Methods: This is analytical observational retrospective cohort study.
Adult patients who were started on meropenem were included. The primary
outcome was a composite of in-hospital mortality, fever recurrence and intensive care unit admission. Univariate and multivariate analysis were performed to determine associations with the primary outcome, controlling for
confounding variables.

Results: One hundred twenty-eight patients (median age 51 years, interquartile range 34–63.5) were included, with lower occurrence of the primary
outcome (21.2% vs 41.1%; P = 0.041) and intensive care unit admission
(3% vs 16.8%; P = 0.044) in the early de-escalation group. In the multivariate analysis, de-escalation lowered the primary outcome risk (odds ratio
[OR], 0.28; 95% confidence interval [95% CI], 0.09–0.85; P = 0.025),
while neutropenia >10 days increased the risk (OR, 1.20; 95% CI,
1.09–1.33; P < 0.001). A lower risk was found in the autologous transplant
(OR, 0.12; 95% CI, 0.21–0.63; P = 0.013) and the consolidation groups
(OR, 0.09; 95% CI, 0.01–0.81; P = 0.031) compared to the induction
chemotherapy group.

Conclusions: In hematologic malignancy-associated febrile neutropenia
patients who remain hemodynamically stable and without documented infectious foci, the early de-escalation from carbapenem may be associated
with better clinical outcomes. Prospective and multicenter studies are
needed to confirm these findings.