Setting and study period

This study was conducted at the “Seràgnoli” Institute of Hematology (IH) at IRCCS Azienda Ospedaliero-Universitaria di Bologna, a tertiary hospital of Northern Italy.

The IH includes three different units located on three separate levels. The outbreak was observed at the third floor of the building, which can host 21 patients, mostly affected by multiple myeloma (MM) and acute myeloid or lymphoblastic leukemia (AML and ALL). As shown by the floor plan in Figure 1, rooms at the third floor are composed as following: 8 double rooms (sharing bathrooms two by two), 1 triple room and 1 single room, both with private bathrooms.

Figure 1.

Third floor plan of the hematology department.

Figure 1. Long description

A floor plan of the hematology department on the third floor. The plan includes various rooms and areas labeled with their respective functions. The layout features multiple patient rooms, each equipped with necessary medical facilities. There are also several common areas, including corridors, nursing stations, and utility rooms. The plan is detailed with dimensions and specific labels for each room and area, providing a comprehensive view of the department’s structure.

The overall time frame of cases associated with the outbreak ranged from February 13, 2024, to September 16, 2024, during which microbiological assessments and epidemiological investigations were conducted to confirm the correlation between cases. Clinical follow-up of the patients involved in the outbreak and epidemiological data regarding microbiological surveillance and colonization status of patients hospitalized in the IH were collected until the 28th of September 2025.

Patient consent statement

The study was conducted according to the declaration of Helsinki and Good Clinical Practice guidelines and was approved by the local Ethics Committee (n° 894/2021/Oss/AOUBo). Informed consent was obtained contacting patients via email or phone call. In case it was not possible (eg, deceased or untraceable patients), the informed consent was waived considering the observational nature of the study and data were collected anonymously, meaning that possible patient identifiers were removed.

IPC protocol, antimicrobial stewardship and microbiological screening

Standard IPC measures in the IH included mandatory compliance with hand hygiene, the use of dedicated patient-care equipment and strict central venous catheter care bundles. Standard environmental protocols involved daily high-level chlorine-based disinfection and a total restriction on live plants or flowers.

In light of the vulnerability of the hematological population and the structural limitations of the building, a dedicated infectious disease consultation service was established in 2020 to conduct daily clinical rounds and drive antimicrobial stewardship programs, focusing on the optimization of therapies and prophylaxis in the IH.

Standard surveillance protocol in the IH, before the study period, consisted of a weekly culture-based screening for rectal CPE carriage, using a selective media plate (Kima Meus, Padova, Italy), followed by confirmation and identification of the enzymatic mechanism in case of positive culture performed by immunoassay NG-Test Carba5 (NG-Biotech, Guipry, France).

Upon detection of CPE colonization or infection, IPC protocol included patient relocation in a single room with private bathroom, whenever possible, otherwise cohorting with other patients colonized or infected by the same multi-drug resistant organism. In case the two previous options were not available, functional isolation was adopted: a peripheral, delimited area within the room less exposed to the transit of people and materials had to be identified and dedicated exclusively to the colonized or infected patient. Contact precautions were required for healthcare staff, including the use of disposable gloves and gowns, hand disinfection and the use of dedicated medical equipment.

Outbreak identification and interventions

Between February 13, 2024 and March 6, 2024 four patients hospitalized at the 3rd floor of the IH, were diagnosed with oxa-48 infection or colonization (3 bloodstream infection and 1 asymptomatic bacteriuria) despite initial negative surveillance rectal swabs. Thus, the following procedures were implemented:

• – targeted inspections by the Hospital Infection Prevention and Control Sector were carried out across three different days (March 6, 7, and 13, 2024) to verify adherence to prevention measures and the correct implementation of cohort isolation. Key infection control measures were enhanced, targeting all personnel and visitors entering the IH. Recommendations focused on ensuring proper hand hygiene, avoiding jewelry to facilitate wrist disinfection, correctly wearing personal protective equipment, supervising the access and behavior of external contractors or consultants, and maintaining rigorous traceability for medical device sanitization. Particular attention was paid to environmental sanitation, especially in the case of shared restrooms and common areas. Furthermore, patients were advised to restrict movement outside their room/unit as much as possible, except for essential diagnostic or therapeutic procedures.

• – Rectal colonization screening was shifted from culture to molecular-based using Xpert Carba-R assay (Cepheid, Sunnyvale, USA) directly on the rectal swab samples and when positive, culture isolation and susceptibility testing were then performed. This approach was chosen due to a suspected false-negative screening result with cultural methods alone. Specifically, the susceptibility profiles of the BSI strains (see Supplementary Table 1) exhibited carbapenem minimum inhibitory concentrations within the susceptible range. Weekly molecular-based screening was first introduced at the 3^rd floor of the IH (starting from March 6th), then implemented in the whole IH (starting from May 7th).

Definitions

We defined as colonization the detection of OXA-48 like–EC from non-sterile sites (urine, sputum, rectal mucosa) in the absence of infection signs or symptoms.

We defined as infection sustained by OXA-48 like-EC the isolation of the same organism from a normally sterile site, or non-sterile site with clinical correlation, accompanied by systemic or organ-specific signs consistent with infection.

The index case of the outbreak was defined as the first patient showing either colonization or infection by OXA-48 like-EC and secondary cases were defined as subsequent patients who tested positive.

Microbiological diagnostic methods

Species identification was carried out using Matrix-Assisted Laser Desorption/Ionization Time-of-Flight mass spectrometry (Maldi Biotyper Sirius system, Bruker Daltonics, Bremen, Germany). Seven bacterial strains underwent WGS for an in-depth characterization of their genomic profile, including antimicrobial resistance genes and virulence factors. Genomic DNA (gDNA) was extracted using the DNeasy Blood and Tissue Kit (Qiagen, Basel, Switzerland) starting from bacterial cultures obtained by incubating the strains of interest on Tryptic Soy Agar plates (Liofilchem, Italy) at 37°C for 18–24 hours. The resulting gDNA was subsequently purified with AMPure XP beads (Beckman Coulter, USA) and quantified using the Qubit™ dsDNA BR Assay Kit (Thermo Fisher Scientific, USA).

Sequencing libraries were generated using the DNA Prep Library Preparation Kit (Illumina, California, USA) and sequenced on the Illumina MiSeq platform (Illumina, San Diego, CA, USA) according to the manufacturer’s instructions.

Quality trimming of the paired-end reads was performed using Trim Galore(ref: https://github.com/FelixKrueger/TrimGalore), setting a minimum length of 50 bp, a minimum quality score of 20, and adapter auto-detection. The quality of the processed reads was then evaluated with FastQC(ref: https://github.com/s-andrews/FastQC). Genome assembly was conducted using SPAdes v4.0.0. The Sequence Type (ST) of the isolates was determined via the MLST tool(ref: https://github.com/tseemann/mlst), which compares assembled genomes with traditional typing schemes from PubMLST. ^{Reference Jolley and Maiden10} Annotation of the assembled genomes was performed with Prokka v1.14.6. Serological types (serotype), FIM, and CH were identified using the SeroTypeFinder, FimTyper, and CHTyper tools, available through the Center for Genomic Epidemiology(ref: https://genomicepidemiology.org/). Resistance genes and virulence factors were identified using the AMRFinderPlus program. ^{Reference Feldgarden, Brover and Gonzalez-Escalona11} Plasmid identification was carried out with PlasmidFinder. ^{Reference Carattoli, Zankari and Garciá-Fernández12}

The genomic correlation of the isolates was evaluated by comparing them with all E. coli ST410 genomes (n = 1733) available in the Enterobase database. Phylogenetic analysis was conducted using Parsnp from the Harvest suite, ^{Reference Treangen, Ondov, Koren and Phillippy13} including the 1733 genomes and the 7 isolates, using CAVI-FOS9 as the reference genome. The phylogenetic tree was visualized using iTOL. Finally, the clonality of the isolates involved in this outbreak was assessed using Snippy v.4.6.0.