Do we overtreat patients with presumed neutropenic sepsis?

PURPOSE: Many aspects of the management of neutropenic sepsis remain controversial. These include the choice of empiric antibiotic, the duration of antibiotic therapy and the possibility that very low-risk cases may be managed safely with oral rather than intravenous therapy. STUDY DESIGN: Retrospective cohort study conducted in a district general hospital serving a population of 148 000 in south west Scotland. RESULTS: Fifty one patients with cancer, whose neutrophil count was less than 1.0×109/L within 21 days of their last chemotherapy, were admitted as a medical emergency in 2019. All received antibiotic because of presumed neutropenic sepsis. A total of 4 patients had positive blood cultures (group 1), 12 patients had a clinical focus of infection but no clear pathogen (group 2), while 35 patients had neither (group 3). Group 3 patients were more likely to have a solid tumour, less likely to be febrile, had shorter time to neutrophil recovery and higher Multinational Association of Supportive Care in Cancer scores, though not all of these comparisons achieved statistical significance. Median intravenous plus oral antibiotic duration in group 3 patients was 9 days with median hospital stay of 7 days, raising the possibility of overtreatment. Retrospectively, 23 (66%) group 3 patients had MASSC Risk Index greater than 21 suggesting they were at low risk of complications. CONCLUSIONS: It seems likely that many low-risk neutropenic cancer patients with solid tumours could be managed as effectively and as safely with shorter courses of antibiotic, with oral rather than intravenous antibiotic, as outpatients rather than inpatients and with an overall positive impact on antimicrobial stewardship.