1 reply to this topic

[_creeks_]

Hello everyone,

I work as a lab technician in a goat cheese manufacturing facility, where we perform routine microbiological controls on both intermediate and finished products. We use 3M Petrifilm Coliform Count plates for in-house testing.

This week, we obtained a coliform count of approximately 1700 cfu/g on a finished, MAP-packed goat cheese product made from pasteurized milk. To rule out errors, I re-tested a counter sample the next day and got a similar result. Confident in our process, we sent a retained sample to an accredited external lab, which reported <10 cfu/g.

This contradiction have left me puzzled. All handling and incubation conditions were properly controlled, and the samples came from the same production lot.

I’d like to ask:

Has anyone else faced such discrepancies between Petrifilm and external lab results?

What factors could explain such a large difference in coliform counts?

How do you usually handle this kind of situation when it comes to internal quality records or non-conformities?

I'd say this happens quite often...
I’d really appreciate your insights or shared experiences.

Thanks!!

[GMO]

Hi, I have never worked in a micro lab but I've audited them.  I've never found an issue specifically between petrifilm and traditional plates.  To my understanding they're both validated processes.

 

I'd say that microbial loading on cheese can be patchy even within the same batch.  Do you have any situations where the cheese is handled, cut or packaged before you test?  If so, my money would be on both results being genuine.  You can get real differences through handling.

 

But the best way to test this would be to do proficiency testing (yourselves and the lab) and check how good your results are.

 

My little caveat to all this though is a bit of a story.  I have directly and indirectly been involved in two lab incidents where faking of results was happening.  One was an external accredited lab where they were sending back fake negatives.  The other was an in house lab where an employee was not testing samples but entering results as <10.  We caught that one because swabs were in the bin with the same volume of nutrient medium still present.

 

It happens and the reason it happens is because lab work is badly paid and high volume.  Perhaps a way to "test" this will be to actually split some samples in your production area (the same cheese) and send half for testing internally and have externally.  Of course there could and will be some differences but you'd expect them to be similar.  The other thing would be to audit your lab and if they have CCTV, see if you can go back to when your specific samples were tested.  

 

Sorry to be suspicious, it's just happened in the UK a few times.