Clinical Outcomes of Pandrug-Resistant Versus Carbapenem-Resistant, Colistin-Susceptible Acinetobacter baumannii Infections: A Retrospective Analysis of a Prospective Multicentre Cohort

Résumé fourni par la source

Background: Pandrug-resistant (PDR) Acinetobacter baumannii represents a major therapeutic challenge in regions where carbapenem-resistant A. baumannii (CRAB) is endemic. Whether the PDR phenotype independently worsens clinical outcomes beyond the effects of disease severity and therapeutic limitations remains uncertain. This study compared the characteristics, management, and outcomes of severe infections caused by PDR and carbapenem-resistant, colistin-susceptible A. baumannii. Methods: We conducted a retrospective analysis of prospectively collected data across 11 tertiary-care hospitals in Greece (February 2022–June 2024). Consecutive adults with bloodstream infection or hospital-acquired/ventilator-associated pneumonia caused by CRAB or PDR A. baumannii were enrolled. The primary outcome was 14-day clinical failure; secondary outcomes included 28-day mortality, microbiological eradication, organ dysfunction, and organ-support-free days. Multivariable logistic and Cox regression analyses, before and after propensity score matching, were performed to adjust for confounding. Results: Among 142 patients, 91 (64%) had PDR and 51 (36%) had carbapenem-resistant, colistin-susceptible infections. Clinical failure occurred in 41% of patients and did not differ significantly between PDR and CRAB infections (39% vs. 45%; p = 0.440). Twenty-eight-day mortality was 33% and 22%, respectively (p = 0.139). After adjustment, the PDR phenotype was not independently associated with clinical failure or mortality. Higher APACHE II score and pneumonia independently predicted clinical failure, whereas sulbactam-containing therapy was associated with lower odds of failure (OR 0.24, 95% CI 0.07–0.79). Older age, higher SOFA score, impaired lactate clearance, and tigecycline-containing therapy independently predicted 28-day mortality. In matched analysis, PDR showed a non-significant upward trend in 28-day mortality (HR 2.36, 95% CI 0.97–5.76; p = 0.059). Conclusions: In severe A. baumannii infections, the PDR phenotype was not an independent determinant of clinical failure or short-term mortality. Patient severity and antimicrobial strategy were major outcome correlates; treatment associations should be interpreted cautiously given the observational design.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.