Abstract
Background
Shorter durations have shown similar clinical outcomes as longer durations for uncomplicated (source-controlled) Gram-negative bloodstream infections (BSI). There is limited data on the outcomes of patients with non-pneumococcal streptococcal BSI receiving shorter durations of therapy compared to usual durations.
Methods
This was a retrospective, multicenter study of adults hospitalized between January 2018 and March 2019 with ≥ 1 blood culture positive for Streptococcus spp. Exposed patients were those who received ≤ 10 days of antibiotics (i.e., short course therapy) and unexposed patients were those who received 11-21 days of antibiotics (i.e., prolonged course therapy). Patients were excluded if they had S. pneumoniae BSI, suspected contamination, did not receive or complete therapy, or treated for > 21 days. The primary outcome was a composite of recurrent bacteremia with the same pathogen, hospital readmission, or all-cause mortality, all within 30 days from completing therapy. The odds of achieving the primary outcome was compared between exposed and unexposed patients using multivariable logistic regression analysis.
Results
A total of 176 patients met eligibility criteria. 35 (20%) received a short course (median 8 days) and 141 (80%) received a prolonged course (median 15 days) of antibiotic therapy. Baseline characteristics were similar between short and long course groups. The most common pathogens were viridans group streptococci (22%) and S. agalactiae (23%). The most common BSI source was skin and soft tissue infection (SSTI) (40%). The primary outcome occurred in 26% (9/35) and 23% (33/141) of patients in the short course and prolonged course groups, respectively (p = 0.774). The proportion of patients in the short course and prolonged course groups who experienced recurrent BSI, hospital readmission, or all-cause mortality were also non-significant. After adjusting for receipt of an infectious diseases consult, Pitt bacteremia score, and SSTI source, the adjusted odds of meeting the composite outcome remained unchanged (aOR 1.41, 95% CI 0.55 – 3.61, p = 0.466).
Table 1. Cohort Characteristics
Table 2. Source/Microbiology
Table 3. Outcomes
Conclusion
Approximately a week of antibiotic therapy may be associated with similar clinical outcomes as longer antibiotics courses in patients with uncomplicated streptococcal BSI.
Disclosures
Kate Dzintars, PharmD, Nothing to disclose Sara E. Cosgrove, MD, MS, Basilea (Individual(s) Involved: Self): Consultant Pranita Tamma, MD, MHS, Nothing to disclose