Search

Abstract Background: Emerging data surrounding the rise in antimicrobial resistance have prompted a shift towards shorter antibiotic durations. Studies show similar clinical outcomes comparing shorter antibiotic courses to longer ones for uncomplicated Gram-negative bloodstream infections (BSI). However, there is a lack of data to inform durations of therapy for Streptococcal BSI. Methods: This was a retrospective cohort study of patients admitted to University of Utah Health with uncomplicated Streptococcal BSI. Inverse probability of treatment weighting (IPTW) was used to estimate the average treatment effects (ATE) of antibiotics administered for 10 days or fewer (short duration) versus more than 10 days (long duration). The primary outcome was a composite of recurrent BSI, all-cause mortality, and readmissions at 30 days from end of therapy. Results: Five hundred patients were screened and 196 were included in the final analysis. The most common sources were skin and soft tissue infections. The median duration in the short and long groups were 8 (IQR, 7–10) and 15 days (IQR, 14–17), respectively. The ATE of short versus long duration of antibiotics was not significant for the composite primary outcome (18% vs 18%; OR = 1.42 [95% CI: 0.57 to 3.53]). Conclusions: We found no appreciable difference in outcomes between patients treated with short versus long antibiotic durations for uncomplicated Streptococcal BSI. Given low absolute rates of mortality and recurrent BSI, along with the lack of evidence indicating a significant difference related to treatment duration, it is reasonable to consider shorter durations. Future research is needed to confirm our findings.

Repeating nasal methicillin-resistant Staphylococcus aureus (MRSA) polymerase chain reactions (PCRs) within 14 days may increase healthcare costs and inform anti-MRSA antibiotic therapy without known benefit. Within an inpatient admission, our retrospective, single-center evaluation found that conversion from negative to positive on repeat nasal MRSA PCR screen was uncommon (2%).