150. Urinary Tract-Associated Gram-Negative Bacteremia: Impact of Treatment Duration

Background Gram-negative bloodstream infections are one of the leading causes of death in the United States. A select number of studies have been conducted evaluating various treatment durations; however, none have specifically focused on urinary sources. The purpose of this study was to compare the effect of short vs. long course of antimicrobial therapy on clinical and microbiological outcomes for urinary tract-associated gram-negative bacteremia (GNB). Methods This was a single-center, retrospective review from January 2016 to October 2018. Subjects were screened using a report of all positive GNB cultures. Hospitalized patients ≥18 years of age were included if they had a bacteremia from a urinary source and received an intravenous or a highly bioavailable oral agent for ≥7 days. Patients were excluded due to pregnancy, incarceration, inappropriate definitive therapy, polymicrobial bacteremia, unaddressed source control issues, or death during the treatment course. Short course (SC) was defined as 7–10 days, while long course (LC) was defined as >10 days. The primary composite outcome of treatment failure included both 30-day all-cause mortality and 90-day recurrence. Secondary outcomes included 30-day re-admission, 90-day mortality, resistance development, and C. difficile infection. Results A total of 207 patients were included: 45 patients received SC and 162 received LC. Both groups were similar at baseline in terms of comorbidities, intensive care unit (ICU) admission, and vasopressor initiation. No statistically significant difference in the primary composite endpoint was observed: 2/45 (4.4%) SC vs. LC 10/162 (6.2%), P = 0.66. There was also no difference in other secondary outcomes. Conclusion Consistent with prior studies, we were unable to find a significant difference in clinical failure rates between SC vs. LC for treatment of urinary tract-associated GNB. Generalizability to more complicated cases including those with inadequate source control may be limited; however, these data add to the body of literature supporting the use of shorter antibiotic durations. Disclosures All authors: No reported disclosures.