Genotypic diversity and antimicrobial resistance phenotype of carbapenem-resistant and carbapenem-susceptible Acinetobacter species isolates.

Abstract

Members of the Acinetobacter baumannii complex are common nosocomial pathogens with propensity for developing carbapenem resistance. Much emphasis is placed on carbapenem-resistant Acinetobacter baumannii (CRAB), although carbapenem-susceptible Acinetobacter baumannii (CSAB) isolates also cause serious infections. Genomic and phenotypic differences between these isolates are analysed in this study. Whole-genome sequencing was performed on a collection of 90 non-duplicate isolates of Acinetobacter spp. isolates. Phenotypic susceptibility results performed routinely were also collected and correlated against presence of antimicrobial resistance (AMR) genes. The included isolates were predominantly A. baumannii (n=77), followed by Acinetobacter nosocomialis (n=8), Acinetobacter pittii (n=2), Acinetobacter ursingii (n=1), Acinetobacter soli (n=1), and Acinetobacter colistiniresistens (n=1). After excluding duplicate isolates, most of the 77 A. baumannii isolates were identified as carbapenem-resistant sequence type (ST) 2 strains (n=43), which were associated with the presence of oxacillinase (OXA)-23 carbapenemases. Comparatively, CSAB were more diverse, comprising various STs. Overall susceptibilities to non-β-lactam antibiotics indicate ​highest susceptibility rates to co-trimoxazole, followed by gentamicin and quinolones. Isolates that were resistant to co-trimoxazole possessed sul1 and/or sul2 genes and isolates resistant to gentamicin had armA and/or ant(2″)-Ia genes. CRAB and CSAB were also distinctly different in terms of resistance to non-β-lactam antibiotics with >90% susceptibility to quinolones, co-trimoxazole, and gentamicin in CSAB isolates, compared to 4.2%, 56.3%, and 33.3% in CRAB. Further stratification was performed based on capsular polysaccharide [K locus (KL)] types. The resistance rates amongst the three most common KL types in carbapenem-resistant ST2 isolates showed considerable variation. For gentamicin, 92.9% (13/14) of KL10, 61.5% (8/13) of KL210, and 45.5% (5/11) of KL2 isolates were resistant. For co-trimoxazole, 0.0% (0/14) of KL10, 92.3% (12/13) of KL210, and 36.4% (4/11) of KL2 isolates were resistant. Overall, the most common types were KL10 (n=15), KL210 (n=14), KL2 (n=11), KL24 (n=6), and KL106 (n=5). The most common types seen in blood isolates were KL210 and KL24 (n=4 each). In summary, CRAB in our institution is dominated by ST2 isolates. There was also strong correlation between the AMR genes present with phenotypic resistance in our collection of isolates. KL type provides further stratification of isolates in terms of invasiveness and AMR.