Key points
- Adding only a few samples each day often creates an incomplete puzzle and slows down the investigation.
- A large sampling campaign from day one should help determine whether contamination is already present at start-up or appears during production.
- Crisis sampling plans, product release procedures and internal organization should be prepared before the first contamination occurs.
This article is also available as a podcast in more than 20 languages. Click the ⚙️ Settings icon on YouTube, then choose Audio Track and Subtitles to watch it in your preferred language.
The result has just come in… Listeria has been detected in a finished product.
Ouch!
How should you organize the investigation to identify the source of contamination as quickly as possible and eliminate it?
We asked Benjamin Yger how to run an effective investigation, quickly get an overall view of the situation and progressively focus actions on the most suspicious areas.
What is the first priority after detecting Listeria contamination in a finished product?
I quickly look for an explanation and a source so that the right actions can be taken.
As long as the source is not understood, the measures remain quite general: more cleaning, stronger chemical concentrations, preventive dismantling, more testing and strong involvement from production, maintenance and quality teams.
These measures may be necessary, but they are difficult to maintain for a long time.
The slower the investigation progresses, the more likely teams are to become exhausted without knowing whether their actions are really targeting the problem.
What can a crisis sampling plan look like?
As an example, I may consider a campaign of around one hundred samples divided into two groups: fifty samples at start-up, after cleaning and with the machines running, then fifty more samples during the production day.
The exact number obviously depends on the production line, the facility and the situation.
The objective is not to systematically use the number one hundred. The goal is to quickly collect enough results to answer two essential questions: is the line clean at start-up, and does contamination appear during production?
Why should you avoid adding only a few extra samples each day?
Because each series provides only a small part of the information.
Five results obtained on Monday may suggest one possible cause. Five more results on Tuesday may suggest another.
But conditions have changed between the two days, making it more difficult to connect the information.
The investigation becomes a puzzle where the pieces arrive one by one, making it difficult to quickly see the complete picture.
I therefore prefer to start, from the first available day, with a specific sampling plan that is as broad as possible.
Where do you start when contamination is detected?
I start by checking the conditions in which the production line restarts after cleaning.
The first samples should help determine whether the equipment starts in good conditions. I run the machines to help release any contamination that may be hidden inside the equipment.
If several positive results appear at this stage, I focus the investigation on the cleaning procedure, equipment accessibility, build-up of organic material, damaged seals or parts, machine design and areas that may require dismantling.
In this case, the contamination is probably already present before production even begins.
What can the timing of positive results tell us?
If results are satisfactory at start-up but become positive during production, I focus the investigation on what happens during production activities.
I look in particular for an incoming flow, a raw material or packaging, operator movements, condensation, equipment that releases contamination only when running, or contamination coming from another area.
This distinction between start-up and production helps reduce the number of possible causes much more quickly.
How far should you go when investigating equipment?
Visible surfaces do not always reveal the source.
A machine may look clean on the outside while still containing material or contamination in an inaccessible area.
When several results point to the same area, I consider dismantling the equipment to be justified.
This can be a complex operation. Some parts are difficult to remove and put back in place. Dismantling can also create adjustment problems when the machine is restarted.
Maintenance teams therefore need to understand why this action is required. Several results pointing in the same direction help show that the intervention is based on a clear hypothesis rather than a random search.
How large should the investigation campaign be?
A large campaign may seem expensive at first, but it can significantly shorten the duration of the crisis.
A step-by-step investigation may last two or three weeks. Some contaminations may even seem to disappear temporarily after stronger cleaning, only to return a few weeks later because the exact contamination source was not eliminated.
A large sampling campaign can sometimes identify the affected area and bring the situation back under control in less than five days.
More analyses are performed at the beginning, but they may prevent several weeks of increased monitoring, excessive use of chemicals, unnecessary dismantling, long product holds and continuous pressure on the teams.
How do you balance production continuity and safety?
This decision depends on the severity of the contamination and the situation.
In some cases, stopping production may be necessary to protect consumers.
In other cases, production may be able to continue, but only with strict safety measures in place.
A hold-and-release process must then be organized, even though this can be difficult for products with a short shelf life.
The finished product testing plan must also be significantly reinforced for the affected production line.
If these safety measures cannot be put in place, production cannot simply continue under normal conditions.
What role can rapid methods play during the investigation?
They can reduce the time between sampling, obtaining the result and taking action.
In routine testing, the choice between a classical method and PCR depends on the budget, whether the company has an internal laboratory, shipping times, the number of analyses and how work is organized with the external laboratory.
PCR is generally more expensive. But during a crisis, the time saved can become very important.
Getting information in around twenty-four hours instead of several days allows teams to adjust actions more quickly and reduce the duration of product holds.
I therefore compare the cost of the method with the total cost of the crisis.
How do you organize the return to normal?
I gradually return to the routine monitoring plan.
Once the situation is stable, the reinforced monitoring plan can be reduced step by step to confirm that the contamination does not return.
You should not suddenly move from a very intensive plan back to the normal routine plan.
The company should be able to explain the criteria used to start the reinforced plan, the different monitoring stages, the conditions for reducing controls and the final return to routine operations.
What can be prepared before a crisis happens?
The exact content of the plan will always depend on the contamination, the production line and the results obtained.
However, the general organization should be prepared in advance.
The factory should know who decides when to move to a hold-and-release process, who performs the sampling, how samples are sent, which laboratory can handle the additional workload, who reviews the results and how production will be secured.
Setting up a hold-and-release process for the first time in the middle of a crisis is particularly difficult.
I prefer to define these procedures in advance and then adapt the specific plan to the real situation.
Thank you to Benjamin for sharing his experience.
If you have any questions, feel free to ask them in the comments… together with the other SuperMicrobiologists, we’ll do our best to answer them.
About Benjamin Yger
Benjamin Yger is an engineer specializing in microbiology and food safety, with around 20 years of experience in the food industry, particularly in fresh and ultra-fresh food production.
Through his company QUALSECAL, he now supports food manufacturers and equipment suppliers with food safety, quality systems and hygienic design. He provides consulting, auditing and training services.