Abstract
BACKGROUND: Limited data suggest that serious infections caused by Enterobacterales with a moderate to high risk of clinically significant AmpC production can be successfully treated with cefepime if the cefepime minimum inhibitory concentration (MIC) is ≤2 µg/mL. However, isolates with a cefepime-susceptible dose-dependent (SDD) MIC of 4-8 µg/mL should receive a carbapenem due to target attainment and extended-spectrum β-lactamase (ESBL) concerns. METHODS: This was a retrospective cohort study of hospitalized patients with E. cloacae, K. aerogenes, or C. freundii bacteremia from January 2015 to March 2022 receiving high-dose cefepime or a carbapenem. Cox regression models were used with incorporation of inverse probability of treatment weighting and time-varying covariates. RESULTS: Of the 315 patients included, 169 received cefepime and 146 received a carbapenem (ertapenem n = 90, meropenem n = 56). Cefepime was not associated with an increased risk of 30-day mortality compared with carbapenem therapy (adjusted hazard ratio [aHR], 1.45; 95% CI, 0.79-2.14), which was consistent for patients with cefepime SDD isolates (aHR, 1.19; 95% CI, 0.52-1.77). Multivariable weighted Cox models identified Pitt bacteremia score >4 (aHR, 1.41; 95% CI, 1.04-1.92), deep infection (aHR, 2.27; 95% CI, 1.21-4.32), and ceftriaxone-resistant AmpC-E (aHR, 1.32; 95% CI, 1.03-1.59) to be independent predictors associated with increased mortality risk, while receipt of prolonged-infusion β-lactam was protective (aHR, 0.67; 95% CI, 0.40-0.89). CONCLUSIONS: Among patients with bacteremia caused by Enterobacterales with moderate to high risk of clinically significant AmpC production, these data demonstrate similar risk of 30-day mortality for high-dose cefepime or a carbapenem as definitive β-lactam therapy.