Genomic Characterization of Carbapenem-Resistant Acinetobacter baumannii (CRAB) in Mechanically Ventilated COVID-19 Patients and Impact of Infection Control Measures on Reducing CRAB Circulation during the Second Wave of the SARS-CoV-2 Pandemic in Milan, Italy

Abstract

COVID-19 has significantly affected hospital infection prevention and control (IPC) practices, especially in intensive care units (ICUs). This frequently caused dissemination of multidrug-resistant organisms (MDROs), including carbapenem-resistant Acinetobacter baumannii (CRAB). Here, we report the management of a CRAB outbreak in a large ICU COVID-19 hub Hospital in Italy, together with retrospective genotypic analysis by whole-genome sequencing (WGS). Bacterial strains obtained from severe COVID-19 mechanically ventilated patients diagnosed with CRAB infection or colonization between October 2020 and May 2021 were analyzed by WGS to assess antimicrobial resistance and virulence genes, along with mobile genetic elements. Phylogenetic analysis in combination with epidemiological data was used to identify putative transmission chains. CRAB infections and colonization were diagnosed in 14/40 (35%) and 26/40 (65%) cases, respectively, with isolation within 48 h from admission in 7 cases (17.5%). All CRAB strains belonged to Pasteur sequence type 2 (ST2) and 5 different Oxford STs and presented bla_OXA-23 gene-carrying Tn2006 transposons. Phylogenetic analysis revealed the existence of four transmission chains inside and among ICUs, circulating mainly between November and January 2021. A tailored IPC strategy was composed of a 5-point bundle, including ICU modules' temporary conversion to CRAB-ICUs and dynamic reopening, with limited impact on ICU admission rate. After its implementation, no CRAB transmission chains were detected. Our study underlies the potentiality of integrating classical epidemiological studies with genomic investigation to identify transmission routes during outbreaks, which could represent a valuable tool to ensure IPC strategies and prevent the spread of MDROs. IMPORTANCE Infection prevention and control (IPC) practices are of paramount importance for preventing the spread of multidrug-resistant organisms (MDROs) in hospitals, especially in the intensive care unit (ICU). Whole-genome sequencing (WGS) is seen as a promising tool for IPC, but its employment is currently still limited. COVID-19 pandemics have posed dramatic challenges in IPC practices, causing worldwide several outbreaks of MDROs, including carbapenem-resistant Acinetobacter baumannii (CRAB). We present the management of a CRAB outbreak in a large ICU COVID-19 hub hospital in Italy using a tailored IPC strategy that allowed us to contain CRAB transmission while preventing ICU closure during a critical pandemic period. The analysis of clinical and epidemiological data coupled with retrospective genotypic analysis by WGS identified different putative transmission chains and confirmed the effectiveness of the IPC strategy implemented. This could be a promising approach for future IPC strategies.

Keywords: Acinetobacter baumannii CRAB; genomic surveillance; infection prevention and control (IPC); intensive care unit (ICU); phylogenetic analysis; whole-genome sequencing (WGS).

FIG 1

ICU_MILANO stay of CRAB patients (n = 40) from admission to discharge, including information on hospital of provenance, ICU module, and day of CRAB isolation.

FIG 2

Estimated maximum likelihood phylogenetic analysis of Acinetobacter baumannii isolates (n = 40) in an ICU COVID-19 hub hospital in Italy. The maximum likelihood was inferred from a core genome alignment of 3,078,653 bp. The phylogeny was estimated with IqTree using the best-fit model of nucleotide substitution TIM2+F+R3 with 1,000 replicates and fast bootstrapping. The numbers on leaves represent the sample IDs, and bootstrap values higher than 90 are shown on branches. Information regarding the samples were reported: the ICU module (ICU-CRAB module), date of isolation (1 CRAB date), if the positivity appeared at the first surveillance, within 48 h from the entry in the ICU (CRAB⁺ first surveillance), the sequence type (ST), capsular locus (K locus) and lipooligosaccharide outer core (OC locus), and the presence (solid squares) or absence of antimicrobial resistance and virulence genes and mobile genetic elements (MGEs). Core antimicrobial resistance and virulence genes, shared by all strains, are not reported in the figure.

FIG 3

Bayesian reconstruction incorporating the date of first positivity of the 30 Acinetobacter baumannii isolates grouped in transmission chain clusters. The Bayesian method was inferred from a core genome alignment of 3,078,653 bp. The Bayesian phylogeny was estimated with BEAST by running 3 independent chains for 10 million states, using the best-fit model of nucleotide substitution GTR+G4 with a strict molecular clock and an exponential population growth tree prior. The numbers on leaves represent the sample IDs, posterior probabilities of >0.9 are shown on branches. Information regarding the samples were reported: the ICU module (ICU-CRAB module), date of isolation (1 CRAB date), if the positivity appeared at the first surveillance, within 48 h from the entry in the ICU (CRAB⁺ first surveillance), the sequence type (ST), capsular locus (K locus), and lipooligosaccharide outer core (OC locus), and the presence (solid squares) or absence of antimicrobial resistance and virulence genes and mobile genetic elements (MGEs). Core antimicrobial resistance and virulence genes, shared by all strains, are not showed in the figure. Transmission chains are reported numbered from the oldest to the most recent based on the putative age root (with 95% HPD) calculated with Tracer.