Methods: This was a single-center, retrospective cohort study of adult inpatients admitted to an 874-bed academic medical center in Charlotte, NC between September 1, 2014 and August 31, 2017. Patients with a GP-BSI who received appropriate antibiotic therapy with at least one third of their total course administered PO were included. Patients with GP-BSI caused by staphylococcal species were excluded. The primary endpoint was clinical failure in patients receiving high (≥90%) versus low (<90%) BA agents. Secondary endpoints included clinical failure stratified by antibiotic group, bactericidal versus bacteriostatic agents, and organism. Chi-square and Fisher’s exact tests were used to examine clinical failure.
Results: 103 patients were included, 26 in the high BA group and 77 in the low BA group. The median age was 58, 51% were women, 74.8% of patients had streptococcal bacteremia (26.2% S. pneumoniae), with pulmonary being the most common source (30.1%). There were no major differences in baseline demographic and clinical characteristics between groups. The median treatment duration with IV antibiotics was 4 and 5 days in the high and low BA groups, respectively (p=0.12). There was no statistically significant difference in clinical failure in the high versus low BA groups (19% versus 23%, p=0.66, respectively). Clinical failure stratified by antibiotic group, bacteriostatic versus bactericidal agent (OR 1.43, CI 0.26-7.90), and organism also did not yield statistically significant differences.
Conclusion: These data demonstrate similar rates of clinical failure among patients definitively treated with high or low BA agents for GP-BSI. High BA agents such as fluoroquinolones may not be needed for all patients with GP-BSI, where more targeted beta-lactam therapies may be appropriate. Additional prospective studies with larger sample sizes are needed to further validate these conclusions.
J. Sebaaly, None
M. Templin, None
D. Weinrib, None