8 replies to this topic

[AJL]

Hi! 

Hoping for some input and brainstorming. We have a product which gets filled into a container. The micro tests come back consistently as <10cfu/g. But the product ends up going mouldy. 

Is our testing not good enough? How can we make ABSOLUTLEY SURE to catch mould if it is there at the start? How can I possibly predict mould growth over the shelf life?

Frustrated here. 

 

[EagleEye]

Hi AJL,

 

The term "Micro tests" is quite a general one to move forward with our discussion, I afraid.

 

There could be a number of things to consider in this;

1. Nature of the product

2. Processing / any special treatment adopted to reduce the counts.

3. Packing environment.

4. Species of the mould you encounter. ...... and so on (need more context to your specific scenario)

 

Assuming the product is not indented to be sterile, the following are some of the possible reasons of this issue;

 

a. You might be doing the test immediately because of which the minimum number of cells do not present to show up in the test results.

b. You may not be doing the specific test for the organism you are encountering with.

c. Spores are germinated after a long time in the original product which may not be that quickly germinate in the artificial growth medium.

d. Your production environment air could be a source to inseminate the mould seedings to the product container.

 

Start from reviewing the product nature itself to pursue through the other factors.

 

Best of luck

[AJL]

Hi Eagle eye, 

Yes bad habit I have, where I lack detail in the post. 

We can rule out contamination from the environment (I assume?) as we are testing the finished product, thereby we should in theory, be cacthing mould which has contaminated from the environment under processing?

 

We are testing specifically for mould. All results have been <10cfu/g for mould (testing at a dilution ox 10x , plated on petrifilm). 

But, during the shelf life some samples become spoiled. We obviously have trouble predicting which samples will spoil based on initial data as everything is <10. 

So I was wondering if there is maybe a more sensitive test method...because it has to come from somewhere right? 

Is there a pre enrichment or something along those lines to ensure we get all spores and damaged cells in the testing? What is the absolute gold standard for testing - we need to test for all moulds as there does not seem to be a trend as to which type of mould spoils the product. 

Does this help? 

We test the raw materials as well (nothing shows up there). I can't go into more detail about the product itself without divulging confientiality. 

[kingstudruler1]

the problem is probably not your test method.  But it is sampling.  

 

 

if your initial mold level is only 1 colony per 100 g/ml, your chance of finding that one colony is very small doing a single 1:10 dilution.  for this reason, pre enrichment probably not work as well.  

 

my guess is you need to be using a larger sample size

 

if product is a liquid - membrane filtration or other methods to test larger sample sizes could be used.  

 

without knowing matrix / product /etc the best answer is hard to give.    

 

 

I would also search the database, as I recall Charles has posted mountains of information on the subject.  

[AJL]

Hmm, a larger sample size, Yes, possibly we can do that. 

Background info: There are many samples taken of a batch, and these are all sampled and tested. Therefore I was wondering if some if some kind of pre enrichment would help...

But the filtration gave me an idea :D

We have a set up which we use for potable water, could possible be used to make a filtration of the sample. No reason why we can't use this to filter a liquid product, right? 

Great idea. :D

Is there some kind of standard method for this? 

[kingstudruler1]

If you are using what I think you are using for potable water testing, that is exactly what I am talking about.    if your product flow through a 0.45 um filter under vacuum?   it might be useful.  

 

I would reach out to someone like millipore sigma for the exact method for your product.  not sure if they are in Germany, but I would think so.  

 

https://www.sigmaald...bial-filtration

 

example:

https://corn.org/wp-...lter-Method.pdf

 

do you do any testing after initial?  Ie can you detect the issue after a week of shelf life?   Have you tried?   Do you have the time to spare?

 

again, your challenge will be to make sure your sampling / testing is statistically valid for what you are trying to find. 

[AJL]

Thanks King studs ruler. Yeah I checked out the filter(suction cup - ours is the same just a different brand.

We use petrifilm but you can also just take the filter on petrifilm pre moistened with sterile buffer. Thats what we do for potable water testing. :)

It seems like a great idea. 

 

I am imagining:

Take 10g sample, 90 g buffer... homogonise/stomacher... then filter. So the sensitivity would end up being <1cfu/10g instead of being <10cfu/g, right? 

First we have technically only 'tested' 0,1g, but in the suggested method we are testing 10g. 

I love it! 

Can see where you are coming from with the sampling.... Dollars and time, dollars and time. 

We do a 'stress test' for the samples where they are left out at ambient and then tested at EOL. 

Any more input is welcome :D

Thanks 

Edited by AJL, 11 March 2023 - 08:36 PM.

[Charles.C]

Thanks King studs ruler. Yeah I checked out the filter(suction cup - ours is the same just a different brand.

We use petrifilm but you can also just take the filter on petrifilm pre moistened with sterile buffer. Thats what we do for potable water testing. :)

It seems like a great idea. 

 

I am imagining:

Take 10g sample, 90 g buffer... homogonise/stomacher... then filter. So the sensitivity would end up being <1cfu/10g instead of being <10cfu/g, right? 

First we have technically only 'tested' 0,1g, but in the suggested method we are testing 10g. 

I love it! 

Can see where you are coming from with the sampling.... Dollars and time, dollars and time. 

We do a 'stress test' for the samples where they are left out at ambient and then tested at EOL. 

Any more input is welcome :D

Thanks 

Hi AJL,

 

I agree KSR's logic.

This is why membrane approach was developed for water by direct count.

BUT this sensitvity aspect is also why MPN method was developed although likely to involve more manual work as compared to membrane if latter is feasible.

 

The specific shelf life may also restrict yr investigative options, eg 1 week vs 1 year.

 

Fortunately, IM(product)EX, moulds not much problem so not very familiar with this defect. However IIRC, even for routine work, extended incubations are often involved which suggests that low level testing may likely be a challenge. Regardless, the typical preferred procedure IMEX  is to initially start at the beginning of Process and work forwards.

[AJL]

Thanks Charles - would you share more info on MPN technique for moulds? I have only heard about it for E.coli/coliorms... :)

shelf life is in months... and accelerated shelf life trials have their limitations... it has a max temp of 8 degrees.