2244. Clinical Outcomes with Extended Infusion (EI) vs. Intermittent Infusion (II) of Cefepime (FEP), Piperacillin/Tazobactam (TZP), and Meropenem (MEM) in Patients with Gram-Negative (GN) Bacteremia

Abstract

BackgroundThe increase in drug-resistant pathogens has prompted interest in dosing of β-lactams (BLs) via EI. Available data on this practice are conflicting, rarely assess non-critically ill patients or low MIC pathogens, and do not focus on outcomes other than clinical cure or mortality. Further assessment of this practice is warranted.MethodsThis is a retrospective cohort study of adult patients who received FEP, TZP, or MEM for GN bacteremia via INI or EI from 2010 to 2018. Patients were included if the pathogen was susceptible to the target BL and they received study drug within 24 hours of bacteremia onset and continued it for ≥ 48 hours. Patients were excluded if they had a mixed infection or received > 48 hours of combination therapy. Patients were matched 1:1 based on study drug utilized, sepsis severity, ICU status, bacteremia source, and causative pathogen. Outcomes assessed included time to clinical stabilization as well as treatment failure, mortality, length of stay (LOS), and recurrence.Results268 patients (134 matched patients in each group) were included. Median (IQR) age of the cohort was 64 (55–77) years, and 57% were male. Common comorbidities were diabetes (35%) and CKD (26%), and the median (IQR) Charlson Comorbidity Index (CCI) was 3 (1–4). 40% of the population presented with severe sepsis or septic shock, and 42% were in the ICU at infection onset. Baseline characteristics were similar between the two groups except patients receiving EI were older (65.5 (58–78) vs. 61.5 (52–73); P = 0.006 and had higher median CCI (3 (2–4) vs. 2 (1–3); P < 0.001,) while patients in the II group had a higher mean weight (85.5 ± 27.8 vs. 78.8 ± 19.2, P = 0.02.) The most common organisms isolated were E. coli(41%), K. pneumoniae (18%), and P. aeruginosa (13%), and the most common source of infection was the urine (51%). Outcomes are listed in Table 1. EI was associated with decreases in time to defervescence, WBC normalization, and SIRS resolution. Furthermore, EI was associated with a lower incidence of treatment failure and recurrence as well as decreases in LOS and ICU-LOS. There was no difference in mortality.ConclusionThe findings of this analysis highlight the role of EI in BL therapy as an important stewardship strategy to optimize clinical outcomes in all patients with GN bacteremia. Disclosures All authors: No reported disclosures.