Adaptive resistance to cefiderocol in carbapenem-resistant Acinetobacter baumannii (CRAB): Microbiological and clinical issues.

ABSTRACT

Objectives: Determining the best available therapy for carbapenem-resistant Acinetobacter baumannii (CRAB) infections is a challenge. Cefiderocol is an attractive alternative drug effective against many resistance mechanisms in Gram-negative bacteria. However, its place in the treatment of Acinetobacter baumannii infections remains unclear and much debated, with contradictory results. Methods: We describe here the case of a 37-year-old man with ventilator-associated bacteraemic CRAB pneumonia in an intensive care unit. He was initially treated with a combination of colistin and tigecycline, and was then switched onto colistin and cefiderocol. We then used a new accessible protocol to test 30 CRAB isolates (OXA-23/OXA-24/OXA-58/NDM-1) for adaptive resistance to cefiderocol (ARC) after exposure to this drug. Results: After clinical failure with the initial combination, we noted a significant clinical improvement in the patient on the second combination, leading to clinical cure. No ARC was detected in the two OXA-23 case-CRAB isolates. All NDM-1 CRAB isolates were resistant to cefiderocol in standard tests; the OXA-23, OXA-24 and OXA-58 CRAB isolates presented 84.2 %, 50 % and 0 % ARC, respectively. Conclusions: ARC is not routinely assessed for CRAB isolates despite frequently being reported in susceptible isolates (69.2 %). Subpopulations displaying ARC may account for treatment failure, but this hypothesis should be treated with caution in the absence of robust clinical data. The two main findings of this work are that (i) cefiderocol monotherapy should probably not be recommended for OXA-23/24 CRAB infections and (ii) the characterisation of carbapenemases in CRAB strains may be informative for clinical decision-making.