In 2018, the SFOPH recorded 52 human cases of listeriosis, corresponding to a normal annual incidence rate of 0.6 cases/100,000 inhabitants (10). However, during March 6, 2018–July 31, 2018, an increase of L. monocytogenes serotype 4b from 13 human cases was recorded. Whole-genome sequencing (WGS) was performed on these strains by using MiSeq next generation sequencing technology (Illumina, https://www.illumina.com). Sequencing reads were mapped against an mutlilocus sequencing typing (MLST) scheme based on 7 housekeeping genes and a 1,701-locus core genome MLST (cgMLST) scheme by using Ridom SeqSphere+ software version 5.1.0 (11). STs and cluster types (CTs) were determined upon submission to the L. monocytogenes cgMLST Ridom SeqSphere+ server (https://www.cgmlst.org/ncs/schema/690488/).

A cluster was defined as a group of isolates with <10 different alleles between neighboring isolates (9,11). Twelve of 13 isolates were assigned to ST6 CT7448, a unique profile in the database, showed by cluster detection to be closely related. Accordingly, we defined an outbreak case-patient as a patient who had listeriosis and L. monocytogenes ST6 CT7448. An outbreak investigation was initiated by the SFOPH, and patients were contacted to assess food exposures by using a standardized questionnaire. Diagnostic and cantonal laboratories were notified nationwide to ensure rapid submission of L. monocytogenes isolates to the National Reference Centre for Enteropathogenic Bacteria and Listeria for laboratory typing, including WGS. However, the questionnaire-based outbreak investigation did not lead to a suspect food, and the vehicle of infection remained unknown.

Figure 1

Figure 1. Cases of human listeriosis caused by Listeria monocytogenes ST6 CT7488, by week and year, Switzerland, 2018 and 2020. CT, cluster type; ST, sequence type.

In a second wave, onset dates ranged from January 22 to May 26, 2020 (Figure 1). Another 27 cases of infection with L. monocytogenes serotype 4b were recorded; 4 cases were in hospital patients who had underlying conditions. During this period, questionnaire-based data were not available to support a food hypothesis.

Figure 2

Figure 2. Minimum-spanning tree based on cgMLST allelic profiles of 34 human Listeria monocytogenesisolates, 1 food isolate, and 5 environmental isolates, Switzerland. Each circle represents an allelic profile based on...

A total of 22 strains grouped on the basis of WGS in a tight cluster, with the exception of N20–2045, which differed by >8 alleles (Figure 2). This strain was within the cluster definition. However, in absence of supportive epidemiologic data, we were not able to verify whether N20–0245 was truly involved in the outbreak.

Median age of the patients was 81 years (range <1–99 years). More than half of the patients were female (18/34, 53%). Of the 34 human isolates, 30 were from blood samples and 1 each from an abscess, ascites, maternal placenta tissue, or stool sample (Table). One case of perinatal transmission and 10 deaths (29%) were reported.

On April 30, 2020, a cheese manufacturer reported to the cantonal laboratory detection of L. monocytogenes from a sample of soft (brie) cheese made from pasteurized milk. Analysis had been conducted as part of the manufacturer’s routine quality control practices, which are mandatory in Switzerland (Swiss Foodstuffs Act, Article 23). The cheese isolate N20–639 matched the outbreak strain CT by WGS (Table; Figure 2). The cantonal authorities started the tracing of the distribution chain of the dairy. The cheese producer supplied several buyers who provide cheese to retailers throughout Switzerland. The buyers were requested to immediately stop the delivery of the products of this specific producer.

These findings prompted extensive environmental sampling on the production site of the manufacturer. A total of 50 swab specimens from locations, such as vats, cheese harps, skimming devices, sink drains, brushes, scrub sponges, trays, door handles, ripening cellar floors, and walls were obtained. Swabs were incubated in Half Frazer Broth (Bio-Rad, https://www.bio-rad.com) at 30°C for 48 h. L. monocytogenes was detected by real-time PCR with the Assurance Genetic Detection System (Endotell, https://www.endotell.ch) according to the manufacturer’s instructions. To obtain strains for WGS, 5 enriched Half Frazer Broth cultures were streaked on chromogenic Listeria agar plates (Oxoid, Pratteln, Switzerland) and incubated at 37°C for 24 h.

L. monocytogenes was identified in 11 (22%) of 50 environmental samples, and all 5 sequenced isolates matched the outbreak strain CT (Table; Figure 2). These results lead to a recall on May 5, 2020, of 26 items, including brie, sheep and goat cheese, and organic cheeses; production was stopped immediately. The findings were reported to the Epidemic Intelligence Information System for Food and Waterborne Diseases and Zoonoses. After the recall of the implicated products and a public warning issued by the Federal Food Safety and Veterinary Office, 7 cases of listeriosis caused by the outbreak strain were recorded (Figure 1). The last known case caused by this outbreak strain was sampled on May 20, 2020, and reported to SFOPH on May 25, 2020. Sequence data have been deposited in the National Center for Biotechnology Information (Bethesda, MD, USA) BioSample database under project no. PRJNA640586. We provide accession numbers (Table).

This prolonged outbreak of L. monocytogenes 4b ST6 CT7448 caused 34 laboratory-confirmed listeriosis cases and 10 deaths. The outbreak investigation is an example of successful collaboration between laboratories and food safety and public health authorities to determine sources of contamination and reconstruct outbreak development. The results of the investigation implicated a cheese dairy with sanitation shortcomings and persisting environmental contamination throughout the production site. Isolation and WGS typing of L. monocytogenes from a quality-control cheese sample provided crucial information that enabled identification of the origin of contamination. WGS played a key role in showing close relatedness between the isolates from the cheese item and from the environment, and in linking the listeriosis cases from 2018 to the 2020 outbreak.

This outbreak highlights the risk for recontamination of pasteurized cheese products during manufacturing and emphasizes the need for routine sampling of products, manufacturing equipment, and the production environment. Routine quality controls should include WGS typing of environmental L. monocytogenes isolates to enable early recognition of potential food contamination and to ultimately mitigate the risk for listeriosis.