15 replies to this topic

[abbie]

hi all,
We are still using the conventional method in our company. I would like to know if you have to count everything that appears on the plate including the tiny specks.
thanks

[Jean]

Dear Abbie,



The conventional TPC have limitations like some colonies may be too small to see or even it may be difficult to differentiate debris and bacteria. Especially if pour plate technique is used it becomes difficult to differentiate between the debris from the colonies.

The experienced microbiologist can distinguish the small round (dot), uniform bacterial colony from the irregular specks which may appear at times. If you know the morphology of the colonies this would be help. At times it may be difficult to estimate the counts on the food particles (if present). This can result in wrong readings and sometimes in false positives or negatives.



If you could elucidate more on the issue may be I can help you better. What method do you use for plating?

[Charles.C]

Dear abbie,

This is often more of a problem IMEX at the "first" dilutions in which case it's inconvenience will depend on yr actual situation as requested by Jean. If you are mainly using 10(-3) up etc, shud be less confusing ?

TPC measurements (like many micro. numerical evaluations) are typically / sadly subject to quite large confidence intervals. This can be a problem where specifications are involved.

Rgds / Charles.C

[abbie]

thanks for answering. I'm using pour plate method, using first dilution. The irregular shapes that are underneath a bigger colony is the problem. They appear like stains to me, so I'm not really sure, but I count it anyway. (should i still count them?)

[Charles.C]

Dear abbie,

A little related info. please -

product = ?

typical result = ?, eg 1-10. 100-300

specification max. TPC/g = ?

Rgds / Charles.C

[abbie]

hi charles,

Cocoa powder is our product. 10,000 max. I usally get around 7000 tpc/g. However, I've seen other companies having 5,000 as their max. I'm wondering if its due to my counting. Btw, would you have a sample haccp for our product. thanks

[Charles.C]

Dear Abbie,

OK, thks, I understand the purpose of yr question.

Frankly, I don’t think the two limits of 5000 and 10000 are sig. very different in bacterial terms unless you hv a specific customer factor.

Regret not really competent to comment on the visual part of yr query (usually handling much higher bacteria levels so no similar problem) but I can on the numerical aspect.

IMO, the errors due to sampling, procedure and evaluation will often cause variations in the result of the size you refer. If 2 labs are consistently different for similar products, suggests a method difference of course. The text books all strongly state that large variations can come from small basic procedure differences as you probably know already. For example, I hv experienced changes in personnel giving such effects and failed to find the reason. Similarly, IMEX, Petrifilm plates tend to give lower results but this may be my procedure/product related or whatever, never found out. Precise answers probably require standardised validation procedures.

The most usable method rather depends on yr typical range of results obtained, eg all yr results are 6000 – 8000 or 5000 – 9000 ? (my guess the latter, or wider )
Regarding yr procedure, I presume you are using a routine style, eg 1ml inoculum from a 25g/225 ml origin. This indicates a tpc of 7000 wud be 700 cols on 10(-1). This exceeds the rec. range for counting (eg 25-250) and could in itself give low results due to crowding although it’s not massively in excess.
Have you tried 10(-2) which wud give 70 cols and should hv less identification problems (and less counting !) ?
Additionally the reliability of the procedure can be judged by the ratio of the average results for 10(-1) and 10(-2) as per standard texts.

I appreciate you are probably trying to minimise workload. If you are using shortcuts like <1ml inoculum and/or only 1 dilution, this may demand a careful look at the reliability of the results, especially if comparing to other people or standards. On the other hand, if it's for in-house guideline use only, maybe less critical.

Yr query about HACCP. There is a plan posted on this forum (2006) ( http://www.ifsqn.com...?showtopic=3671 ) but unfortunately the link has now failed. Don’t know if anyone has a copy they cud repost here. Please, please.

Maybe someone with similar product bacteria levels can answer the visual query bit ?

Rgds / Charles.C

[Jean]

thanks for answering. I'm using pour plate method, using first dilution. The irregular shapes that are underneath a bigger colony is the problem. They appear like stains to me, so I'm not really sure, but I count it anyway. (should i still count them?)

Charles has given you some very good suggestions and therefore I will just point out on the spreaders which may be the reason for the tiny spots as you have mentioned.

Colonies which spread may form a chain of colonies which are not separated, a growth from the bottom of the plate and agar or even a film like is formed on the surface of the agar or on the edges. In such a case if only one chain exists from a same clump then count it as one, but if different chains originate from different sources then count each source as one colony. Then combine the spreader to the individual colony counts. I would also suggest you to use dilution factor 10^-2 and duplicates.

[Jean]

Dear Abbie,



All pinpoint size colonies should be counted (if there is no spreader). If you doubt objects in the plate as colonies then use a magnifying glass to distinguish food particles / debris from the colonies. Sometimes the scratches on the plates also affect the reading. If there are a lot of food particles which may interfere with the reading, then you can add TTC to the agar medium prior to pouring. The bacterial colonies will appear red in the plate with TTC, for this again you need to have a plate without TTC to ensure TTC does not have any effect on the counts.



Proper shaking or mixing of the samples and initial dilution may help in uniformly distributing the clumps of bacteria. Also avoid stacking too many plates in the incubators. Excess humidity in the incubator can leader to spreaders therefore adjust the ventilation & air circulation.



By increasing the dilution to 10^-2 & 10^-3 will help you to interpret the results better.

[abbie]

Dearest Charles and Jean,

I think you've answered all my boring questions about microbiology. I would suggest using 10(-3) and using TTC. I think that would give a big difference.

.... also, Charles, I've read your discussion reg. the cocoa powder haccp.(i tried downloading the attachment myself ) I hope i can get your opinion on that next time.

thanks!
abbie

[Charles.C]

Dear Abbie,

I think you've answered all my boring questions about microbiology.

Don't worry. Yr questions are very relevant, not boring at all. I don't like to think how much time I've spent trying to understand why my results for what are claimed to be standard measurements don't agree with other peoples. The most common reason IMEX is a difference in procedures, often because different countries / companies prefer particular methods. Another example is the effect of incubator temperatures on plate counts. It is amazing how many different temperatures are practically in use and this parameter is often not reported at all.
If you are really interested there is a long thread here somewhere on the accuracy of microbiological measurements.

Rgds / Charles.C

[cazyncymru]

thanks for answering. I'm using pour plate method, using first dilution. The irregular shapes that are underneath a bigger colony is the problem. They appear like stains to me, so I'm not really sure, but I count it anyway. (should i still count them?)

So your using a 1/10th Dilution? (first dilution)

hi charles,

Cocoa powder is our product. 10,000 max. I usally get around 7000 tpc/g. However, I've seen other companies having 5,000 as their max. I'm wondering if its due to my counting. Btw, would you have a sample haccp for our product. thanks

If your counting 7000 colonies on a 1/10th Dilution then you really need to be doing a series of dilutions, going down to 1/100000 if needs be.

[abbie]

hi caz and charles
thanks for the comment. There really is a need to change our procedure as it greatly affects the accuracy of the test.(we may be having problems with the cocoa powder) Charles is right about having the routine. We use the first dilution for everything including the cocoa powder. However, due to the difficulty, we'll be using higher dilutions as everybody suggested. thanks thanks.

....

If you are really interested there is a long thread here somewhere on the accuracy of microbiological measurements

charles, I will try to look for that. thanks again.

abbie

[mherz]

Hi,

Thanx for those informations.

I have a question to post also regarding the TPC count.

We are producing PET sheets rolls for food packaging materials.
We are supplying it to our customers who will be thermoforming,laminating the PET Sheet rolls.

My question is what is the acceptable TPC count limit for our product considering the succeding process that it will pass through.

We have just subjected our pallets and cardboard for microbiological testing with a result of 6 cfu/26cm2 and 3cfu/cm2. The pet sheet rolls are totally wrap with cling film then it will be palletized with cardboard on top and bottome to protect the rolls for edge damage while on transit.

hope somebody can help me here.

[siju]

We have got a new media for Total plate count test for mineral water .On the media container they are mentioned that, use for surface inoculation of aerobic microbial count. What is meant by Surface inocculation? I used to carry out Pour plate method with another media but the new media from Scharlau,Spain put me in trouble. Can any body tell me what is Surface inoculation ? Is it Spread plate technique?



Thanks,

Sijo joseph,
India

[Charles.C]

Dear siju,

Apparently yes.

 surface inoculation.png   23.74KB   6 downloads

http://www.vphcap.or...hapter3.pdf.pdf

(3.5.2)

Rgds / Charles.C