Third-generation cephalosporin use is frequently non-guideline-concordant in severe community-acquired pneumonia: Findings from a French critical care cohort

Abstract

Background and objectives Severe community-acquired pneumonia (CAP) remains a major cause of morbidity and mortality, requiring prompt empirical antibiotic therapy. The choice between amoxicillin/clavulanic acid (AMC) and third-generation cephalosporin (3GC) as first-line beta-lactam therapy in the ICU is still debated. To assess AMC and 3GC susceptibility in severe CAP or community-acquired aspiration pneumonia (CAAP) caused by Streptococcus pneumoniae (SP) , Haemophilus influenzae (HI) and/or Staphylococcus aureus (SA) , and to evaluate the appropriateness of empirical antibiotic therapy. Methodology We conducted a single-center retrospective study including patients admitted to the ICU between 01/01/2018 and 30/11/2022 for severe CAP/CAAP with microbiological documentation of at least one of the targeted pathogens. Clinical, microbiological, therapeutic, and outcome data were collected. The primary endpoint was AMC and 3GC susceptibility rates (expressed as percentage and [95% confidence intervals]). Principal findings Among 104 included patients (median age 64 [48–71] years; 67% male), 60 (57.7%) had CAAP. AMC susceptibility rates were 81.5% [61.9–93.7], 78.0% [62.4–89.4], and 100% [92.6–100.0], while 3GC susceptibility rates were 96.3% [81.0–99.9], 95.1% [83.5–99.4], and 100% [92.6–100.0], for SP , HI , and SA , respectively. AMC was the empirical treatment in 59% of cases, and this choice was microbiologically appropriate in 86.9% of these prescriptions. Conversely, 3GC were prescribed empirically in 20% of cases, but this broad-spectrum choice was unjustified in 80% of those situations. Antibiotic de-escalation was performed in 28% of cases. Conclusion In this single-center retrospective cohort, AMC appears to be a relevant empirical option for severe CAP/CAAP in the ICU, while 3GCs seem frequently overused. These findings required prospective multicenter validation before practice change can be recommended.