Objectives: The use of extended infusions of cefepime has been shown to optimize pharmacokinetic and pharmacodynamic parameters. Optimization improves clinical outcomes in adults with infections secondary to gram-negative pathogens, but there are no studies demonstrating benefit in pediatric patients. The objective of this study was to compare clinical outcomes in pediatric patients receiving extended and traditional infusion cefepime.

Methods: This was a retrospective cohort study of patients aged 30 days to 17 years receiving extended versus traditional infusion cefepime at a tertiary care children’s hospital from January 1, 2007 to April 30, 2016. Patients were included if they received cefepime for at least 48 hours for a susceptible gram-negative bacteremia with no concomitant gram-positive or fungal infections. Patients were excluded if they received both regimens, did not receive active antimicrobial therapy within 24 hours of the first positive blood culture, or received more than 24 hours of empiric therapy with a beta-lactam other than cefepime. The primary outcome was a composite clinical outcome comprised of infection-related mortality within 14 days of antibiotic initiation, bacteremic relapse with the same organism within 30 days of culture clearance, and treatment failure. Treatment failure was defined by lack of defervescence, normalization of white blood cell count, or negative follow-up blood cultures. Patients who failed to meet the criteria for defervescence, WBC normalization, or culture clearance or any single component when no other information was available were considered treatment failures.

Results: Sixty-seven patients were included; 21 (31.3%) received extended infusion, and 46 (68.7%) received traditional infusion. Traditional infusion patients were significantly younger and smaller than extended infusion patients with a median (IQR) age of 0.5 (0.2-2) vs. 7 (0.8-11.5) years (p = 0.05) and weight of 6.2 (2.4-13.1) vs. 23.7 (8.2-44.2) kilograms (p<0.05). The most common pathogens within the extended infusion group were Enterobacter cloacae (23.8%) and Klebsiella oxytoca (23.8%), while Escherichia coli was the most common within the traditional infusion group (28.2%). There was no difference in MICs between groups, as the most common MICs were ≤1 mcg/mL in both extended (95.2%) and traditional (87.0%) infusion (p = 0.801). Clearance of bacteremia within 24 hours was observed in 8 (38.1%) patients receiving extended infusion versus 12 (26.1%) patients receiving traditional infusion (p = 0.392). The primary outcome occurred in 2 extended (9.5%) and 3 traditional (6.5%) infusion patients (p = 0.645). Two (9.5%) adverse effects were reported with extended infusion compared to 3 (6.5%) adverse effects with traditional infusion (p = 0.645).

Conclusions: Use of extended infusion cefepime in pediatric patients with documented gram-negative bacteremia led to similar clinical outcomes as traditional infusion in this cohort of children with low bacterial MICs. Larger studies including patients that are infected with organisms with higher MICs may be more likely to detect significant difference in outcomes.