Low APC, High Coliform Count in Pork Dumplings

Hello! 

I'm having an APC/CC problem. The company I'm with produces Chinese pork dumplings, six different flavors. In the Pork and Cabbage dumplings we produced on 06/13/2018, I got coliform counts higher than the APC counts. This occured consistently as I retested the product from the same production date. Does anyone know how I might be getting coliform counts higher than the APC counts? I've included my data and would appreciate some intel. 

Also, just to note--the first three entries on the document I incubated APC at 96-100 degrees F and CC at 84-88 degrees F. For the last test date, I adjusted the APC incubator to 93-97 degrees F, and the CC incubator stayed the same. 

Thanks! 

Beth Estrada 

Production Date: 06/13/2018

Flavor: Pork and Cabbage

(Re)Test Date 1: 08/06/2018

APC 1: 130 (cfu/g)

CC 1: 700 (cfu/g)

(Re)Test Date 2: 08/08/2018

APC 2: 150

CC 2: 620

(Re)Test Date 3: 08/20/2018

APC 3: 140

CC 3: 180

(Re)Test Date 4: 08/27/2018

APC 4: 380

CC 4: 780

I assume you're ding a 1:10 dilution onto the plates? (e.g. multiply count by ten to get cfu/g)

What methodology are you using for each test? Specifically the coliform, 3 tube? petrifilm? automated counter? How many days incubated? REALLY need to know the exact methodology used for each to start guessing why they would report differently.

Could simply come down to having organisms that are growing better on the coliform media vs. the APC media. Or it could be the confidence intervals playing with you. Keep in mind that with quantitative micro really anytime you're within 1 log it's basically the same number. As an example, the 3-tube 95% confidence intervals for a result of 460 is anywhere from 90-2000MPN.

Hi bethre,

(I amended yr data slightly for convenience)

A few observations in addition to 3F's Post.

How do yr reported data compare to the usual results for other Production dates of Pork Dumplings/Other Flavours, eg do all show CC > APC ?

From the APC values i assume this is a cooked RTE product ?

If so, regardless of the APC count, the Coliform values look "high". (see my first query).

As a general comment, one sample/datum of anything is often insufficient for Predictions unless homogeneity is assured. (4 retests does have significance).

I anticipate you are using Violet Red Bile Agar for Coliform count.

IMEX (not pork dumplings) one needs to (at least initially) confirm (by sampling the plates) that the colonies are Coliforms when using VRBA.

I hope you did duplicate plates for the APC/CC counts ?. If so can you give the raw data ? (valuable to see technique consistency)

A confirmed APC result of 130cfu/g with a Coliform count of 700cfu/g would be unusual (rare) for a cooked item IMEX.

I assume you're ding a 1:10 dilution onto the plates? (e.g. multiply count by ten to get cfu/g)

 

What methodology are you using for each test? Specifically the coliform, 3 tube? petrifilm? automated counter? How many days incubated? REALLY need to know the exact methodology used for each to start guessing why they would report differently.

 

Could simply come down to having organisms that are growing better on the coliform media vs. the APC media. Or it could be the confidence intervals playing with you. Keep in mind that with quantitative micro really anytime you're within 1 log it's basically the same number. As an example, the 3-tube 95% confidence intervals for a result of 460 is anywhere from 90-2000MPN.

Hi, 

Sorry for the delay in replying! Yes, I'm doing a 1:10 dilution onto Petrifilm plates. I'm using the Rapid Aerobic Count plates and the Coliform Count plates from 3M. I weigh 25g of the sample and mix it in a 225 ml dilution bag of Butterfield's Buffer (also 3M) using a stomacher. Then I let the solution settle for a few minutes and pipette 1 ml onto each plate. I always change the pipette tips between samples, but not between different plate types. I thought I might just be seeing a lot of bubbles in the CC plates, so I put an opened vial of Butterfield's Buffer in the CC incubator to change the humidity (they told me to do that at my last job) but I'm still seeing a lot of bubbles, and they look like colonies. 

I should also mention that on most (possibly all) of the APC plates holding samples from this production date, there have been some spreader colonies. Our rep from 3M said to count each spreader colony as 1 cfu, but if that's not correct, then I'm counting fewer APC colonies than I should be. Is that correct? 

Thanks so much for your help!

Regards, 

Beth 

Hi bethre,

 

(I amended yr data slightly for convenience)

 

A few observations in addition to 3F's Post.

 

How do yr reported data compare to the usual results for other Production dates of Pork Dumplings/Other Flavours, eg do all show CC > APC ?

 

From the APC values i assume this is a cooked RTE product ?

If so, regardless of the APC count, the Coliform values look "high". (see my first query).

 

As a general comment, one sample/datum of anything is often insufficient for Predictions unless homogeneity is assured. (4 retests does have significance).

 

I anticipate you are using Violet Red Bile Agar for Coliform count.

IMEX (not pork dumplings) one needs to (at least initially) confirm (by sampling the plates) that the colonies are Coliforms when using VRBA.

 

I hope you did duplicate plates for the APC/CC counts ?. If so can you give the raw data ? (valuable to see technique consistency)

 

A confirmed APC result of 130cfu/g with a Coliform count of 700cfu/g would be unusual (rare) for a cooked item IMEX.

Hi, 

Thanks for the help! Our dumplings are raw and frozen. I thaw out a sample before cutting and plating. I mix 25g with a 225ml Butterfield's Buffer dilution pouch from 3M, then plate 1 ml of the sample onto Rapid Aerobic Count plates and Coliform Count plates (also from 3M). This is the only sample that has been consistently getting CC>APC. I haven't been doing duplicate plates, as one dilution seems to yield countable cfu/ml, but that's something I should try. 

Thanks again! 

Regards, 

Beth 

Hi, 

 

Thanks for the help! Our dumplings are raw and frozen. I thaw out a sample before cutting and plating. I mix 25g with a 225ml Butterfield's Buffer dilution pouch from 3M, then plate 1 ml of the sample onto Rapid Aerobic Count plates and Coliform Count plates (also from 3M). This is the only sample that has been consistently getting CC>APC. I haven't been doing duplicate plates, as one dilution seems to yield countable cfu/ml, but that's something I should try. 

 

Thanks again! 

 

Regards, 

 

Beth 

Hi Beth,

I have no experience with "dumplings" but the plate counts seem unbelievably low for a "raw" item. i guess it may depend on what the nature/composition is.

Yr incubation temps seem slightly lower than IMEX. You omitted to inform how long the incubation lasts ? Hopefully the "Rapid" does not mean 1 hour or something similar.

Frankly i'm surprised you can evaluate plates directly from the first dilution. Normally IMEX such plates are unusable/unreliable due the suspension interfering.

I haven't checked but I expect the 3M system is based on VRBA in which case  you can see my/3Fs previous Coliform comments.

I suggest (if not done already) that  you send  samples of  "normal" product and the atypical one to an outside lab for a cross-check using validated methods (eg maybe not VRBA)(or if VRBA, to be accompanied by a comparison MPN method like BAM).

Regardless, the mystery remains as to why one lot of Product gives atypical data (any Process anomalies ?). But first, i would make sure of the methodology (unless already done of course).

Hi Charles, 

"Rapid" here means 24 +/- 2 hours as opposed to 48.  The CC petrifilm is VBRA based, or at least it says on the Interpretation Guide "contains modified Violet Red Bile nutrients." 

Thanks!!

Beth 

extract from BAM, VRBA Procedure -

Invert solidified plates and incubate 18-24 h at 35°C. Incubate dairy products at 32°C (2). Examine plates under magnifying lens and with illumination. Count purple-red colonies that are 0.5 mm or larger in diameter and surrounded by zone of precipitated bile acids. Plates should have 25-250 colonies.

 

To confirm that the colonies are coliforms, pick at least 10 representative colonies and transfer each to a tube of BGLB broth. Incubate tubes at 35°C. Examine at 24 and 48 h for gas production.

 

NOTE: If gas-positive BGLB tube shows a pellicle, perform Gram stain to ensure that gas production was not due to Gram-positive, lactose-fermenting bacilli.

Determine the number of coliforms per gram by multiplying the number of suspect colonies by percent confirmed in BGLB by dilution factor.

The above precaution if included will presumably, sadly, cancel the "Rapid". :smile:

I have had several prolonged arguments with customers' labs "ignoring" this "control". Potential correction factors of 30-50% were involved.

Thank you Charles! I'll give that a try. 

Also, I was wondering--could it be possible that the Rapid APC plates can't show facultative anaerobic bacteria, but that the CC plates do? 

Thanks! 

Beth 

Hi Beth,

I only just realised that yr APC plates are presumably only incubated for 1-day.

IMO yr first priority is to validate yr data, eg posts 2, 6. Urgently.

Hi Charles, 

Okay, yes I'll see if I can send them out to Michelson Lab soon for validation. Would you like me to follow up with you and let you know how it turns out? 

Thanks! 

Beth Estrada

Hi Charles, 

 

Okay, yes I'll see if I can send them out to Michelson Lab soon for validation. Would you like me to follow up with you and let you know how it turns out? 

 

Thanks! 

 

Beth Estrada

Hi Beth,

Before you send samples out, I suggest you query the lab's specific intended methodologies for APC/Coliform counts.

Apologies for ignorance of yr industry but i have a few queries -

What is the usual reference method for determining APC / Coliform for this range of Products ?

What is yr  typical microbial specification for APC/Coliform for this range of Products ? (I assume similar data across range)

Out of curiosity what is the basic process to produce these flavours ?, eg is there an actual cooking step ?

PS - this document may be of some general interest -

Comparative evaluation of 3M Petrifilm Rapid Aerobic Count,2014.pdf   251.16KB   7 downloads

Hi Charles, 

Sorry for the late response! 

No, there isn't actually a cooking step; we mix raw pork, fat and vegetables and then put the filling through the forming machine with sheets of dough. The forming machine shapes the dumplings and the conveyors carry them to a freezer where they're frozen at about -26 degrees C. From there they go into the packaging room to be bagged. 

Typical APC for the raw dumplings is 2000-8000 using a 1:10 dilution (Butterfield's Buffer). I've been using the 3M Interpretation Guides as a reference for counting CFU/ml. 

Thanks so much for your help! Let me know if there's anything else you'd like to know. 

Have a nice weekend!

Beth Estrada 

Hi Charles, 

 

Sorry for the late response! 

 

No, there isn't actually a cooking step; we mix raw pork, fat and vegetables and then put the filling through the forming machine with sheets of dough. The forming machine shapes the dumplings and the conveyors carry them to a freezer where they're frozen at about -26 degrees C. From there they go into the packaging room to be bagged. 

 

Typical APC for the raw dumplings is 2000-8000 using a 1:10 dilution (Butterfield's Buffer). I've been using the 3M Interpretation Guides as a reference for counting CFU/ml. 

 

Thanks so much for your help! Let me know if there's anything else you'd like to know. 

 

Have a nice weekend!

 

Beth Estrada 

Hi Beth,

Thks for the info.

I deduce this product is a (raw) variation of dim sum, siowmai (and other names). I guess something like -

pork dumplings.png   306.64KB   1 downloads

The batch in yr OP clearly has a highly atypical APC result if the usual raw range is 2000 - 8000 cfu/g. Atypical everywhere in fact !

The only official spec. (Hong Kong) I could find for dim sum was for the RTE cooked version which was -

micro - dim sum.png   52.37KB   1 downloads

(Yr APC range for a Raw finished product seems impressive in respect to the above Guidelines for a Cooked Product).

(Although i noticed in another reference which sampled supermarket RTE frozen (cooked) pork dumplings that the APC results were mainly less than 100 cfu/g)

Looks like cannot say much more until you validate the 3M data. :smile:

Thank you so much Charles! I'll see if I can send out our samples to get validated. 

Thanks again and have a great weekend!