Effectiveness of third-generation cephalosporins or piperacillin compared with cefepime or carbapenems for severe infections caused by wild-type AmpC β-lactamase-producing Enterobacterales: A multi-centre retrospective propensity-weighted study

• Published in: International journal of antimicrobial agents 2023-07, Vol.62 (1), p.106809-106809, Article 106809
• Main Authors: Maillard, Alexis, Delory, Tristan, Bernier, Juliette, Villa, Antoine, Chaibi, Khalil, Escaut, Lélia, Contejean, Adrien, Bercot, Beatrice, Robert, Jérôme, El Alaoui, Fatma, Tankovic, Jacques, Poupet, Hélène, Cuzon, Gaëlle, Lafaurie, Matthieu, Surgers, Laure, Joseph, Adrien, Paccoud, Olivier, Molina, Jean-Michel, Bleibtreu, Alexandre
• Format: Article
• Language: English
• Subjects: AmpC; Bloodstream infections; Enterobacterales; Pneumonia
• ISSN: 0924-8579; 1872-7913
• DOI: 10.1016/j.ijantimicag.2023.106809

•The best therapy for wild-type AmpC β-lactamase-producing Enterobacterales is uncertain.•There was no significant difference in mortality between the study treatment groups: third-generation cephalosporin (3GC), piperacillin ± tazobactam, or cefepime/carbapenem (primary endpoint).•Both 3GC and piperacillin ± tazobactam may increase treatment failure due to AmpC-overproduction (secondary endpoint). The optimal treatment regimen for infections caused by wild-type AmpC β-lactamase-producing Enterobacterales remains controversial. This study compared the outcomes of bloodstream infections (BSI) and pneumonia according to the type of definitive antibiotic therapy: third-generation cephalosporin (3GC), piperacillin ± tazobactam, cefepime or carbapenem. All cases of BSI and pneumonia caused by wild-type AmpC β-lactamase-producing Enterobacterales over 2 years in eight university hospitals were reviewed. Patients who received definitive therapy consisting of either a 3GC (3GC group), piperacillin ± tazobactam (piperacillin group), or cefepime or a carbapenem (reference group) were included in this study. The primary endpoint was 30-day all-cause mortality. The secondary endpoint was treatment failure due to infection by emerging AmpC-overproducing strains. Propensity-score-based models were used to balance confounding factors between groups. In total, 575 patients were included in this study: 302 (52%) with pneumonia and 273 (48%) with BSI. Half (n=271, 47%) received cefepime or a carbapenem as definitive therapy, 120 (21%) received a 3GC, and 184 (32%) received piperacillin ± tazobactam. Compared with the reference group, 30-day mortality was similar in the 3GC [adjusted hazard ratio (aHR) 0.86, 95% confidence interval (CI) 0.57–1.31)] and piperacillin (aHR 1.20, 95% CI 0.86–1.66) groups. The likelihood of treatment failure was higher in the 3GC (aHR 6.81, 95% CI 3.76–12.4) and piperacillin (aHR 3.13, 95% CI 1.69–5.80) groups. The results were similar when stratifying the analysis on pneumonia or BSI. Treatment of included BSI or pneumonia caused by wild-type AmpC β-lactamase-producing Enterobacterales with 3GC or piperacillin ± tazobactam was not associated with higher mortality, but was associated with increased risk of AmpC overproduction leading to treatment failure compared with cefepime or a carbapenem. [Display omitted]