In a new paper published in the International Journal of Food Microbiology, researchers from the Norwegian Institute of Food, Fisheries, and Aquaculture Research (Nofima) reported on a successful “Seek and Destroy” process carried out by a smoked salmon producer implicated in a listeriosis outbreak, describing how the firm identified the source of contamination and controlled Listeria monocytogenes in its facility, and highlighting key takeaways for industry.

The Outbreaks

Between February and October 2022, an outbreak of five listeriosis cases were reported in Norway, followed by a second outbreak of seven cases in 2023. Both outbreaks were traced to cold-smoked salmon produced by the same facility (Producer A).

Whole genome sequencing (WGS) identified the outbreak strain as L. monocytogenes ST121, a subtype known for its persistence in food production environments.

Despite routine cleaning and disinfection, the pathogen continued to appear in finished products. In September 2023, Producer A launched a comprehensive Seek and Destroy process—a structured approach combining increased sampling, intensified deep cleaning, and modifications in equipment and hygiene practices.

The Seek and Destroy Process and Outcomes

Over the course of 11 weeks, 329 samples from food-contact surfaces, non-food-contact areas, and salmon fillets were analyzed. Whole genome sequencing (WGS) was conducted on 21 samples representing different sampling weeks, processing steps, and environmental sites.

During the first week, a test production experiment was performed in which portions of fillets were removed at seven different steps along the processing line and analyzed for L. monocytogenes. Findings suggested that contamination occurred after skinning and salting.

WGS analysis of positive samples collected during week one revealed two main L. monocytogenes ST121 clusters. The biggest cluster, which was likely to be the outbreak strain, included isolates from pooled surfaces from the skinning machine and from the conveyor after the skinning machine, as well as from other product and environmental samples.

The outbreak strain was still present in the facility and was transferred to fillets despite no positive samples from the surfaces of cleaned and disinfected machines, suggesting niches within machines that harbored L. monocytogenes.

The skinning machine emerged as the primary source of contamination during weeks 2–7 of the Seek and Destroy process, when it continued testing positive for L. monocytogenes after intensive deep cleaning and disassembly, including hydrogen peroxide mist treatments and heat exposure.

After the skinning machine was replaced, in weeks 8–11, no L. monocytogenes positive samples were found in salmon sampled after skinning or later in the process.

Nofima investigated the skinning machine to identify harborage sites after it was removed from the facility. The machine’s conveyor belts featured a rubber-like surface over a woven fabric layer, going against hygienic design principles that recommend impermeable, coated belt materials that prevent fluid absorption. Additionally, the mixed-material junction in the machine’s air-filled roller, comprising a plastic core with metal end fittings, also had the typical characteristics of L. monocytogenes harborage sites (i.e., limited accessibility, crevices, and moisture retention).

During the eight months following the Seek and Destroy process, L. monocytogenes was sporadically found in environmental samples, but none of nine typed isolates were part of the outbreak cluster. Additionally, none of the 220 product samples analyzed tested positive for L. monocytogenes.

Lessons for Industry

The investigation highlights several key takeaways for seafood processors:

• Hygienic design is critical: Equipment with porous materials or complex interfaces can harbor pathogens. Conveyor belts and rollers should be made from impermeable, easily cleanable materials.
• Chemical cleaning alone may be insufficient: Even aggressive disinfection of disassembled machines failed to eradicate L. monocytogenes. Heat treatment was effective, but could only be applied to parts that could withstand high temperatures.
• WGS provides important insights: WGS confirmed persistence and linked isolates to outbreaks. However, high costs and limited access to public health sequence data can hinder confirmation of the removal of outbreak clones.
• Standard swabbing can miss niches: Normal sampling with cloth swabs did not detect contamination on suspected parts; however, when Nofima immersed machine parts in Listeria pre-enrichment broth, L. monocytogenes was identified on the skinner machine belts and roller. This method is not suitable for real-world use, but it demonstrated that contamination in harborage sites may evade conventional surface sampling.

Overall, identification and removal of the contaminated skinning machine, combined with strengthened cleaning protocols, ultimately led to the successful removal of the outbreak strain from the facility.