9 replies to this topic

[BALA Prasanna]

Dear Team,

                   Currently we are going with USFDA BAM method for E.coli.

 

For this, we can intertrep the result as <3MPN/g or <10CFC/g.

 

In that results are same like Absent.

 

Can we Justify the result, <3MPN/g consider as a Absent/g.

 

Expertise suggest for this.

[AurW]

Hi Kasi4bala, 

 

I've had the same issue and in fact this is slightly different:

 

the <3MPN/g or <10CFC/g result is in fact a detection limit, usually due to the way the preparation is put onto the plate. in other word, this is the lowest level where contamination can be detected, here E.coli. Strictly speaking, this is not equivalent to an absence test as you could have 5 CFC/g and you will still be compliant with the <10 limit.

​An absence test will let you know the presence or not of an organism but it won't tell you how many are present if you have a positive result.

 

I used to change the results on the certificate of analysis to below the detection limit when I had this type of result to avoid confusion.

Hope this helps.

 

Aurélie 

[Charles.C]

Dear Team,

                   Currently we are going with USFDA BAM method for E.coli.

 

For this, we can intertrep the result as <3MPN/g or <10CFC/g.

 

In that results are same like Absent.

 

Can we Justify the result, <3MPN/g consider as a Absent/g.

 

Expertise suggest for this.

 

Hi KASI,

 

I assume you are referring to generic E.coli.

 

I cannot see any procedural option in BAM for E.coli which offers <10 cfu/g. Please clarify.

 

I presume the <3 refers to the MPN method for a 3x3  setup, ie For 3 tubes each at 0.1, 0.01, and 0.001 g inocula

In fact you can get even more exact by using more tubes. This is detailed in BAM.

 

Assuming the above mix, the interpretation (eg see BAM Table1) of <3 MPN/g is  - "the outcome with all negative tubes is listed as less than the lowest MPN for an outcome with at least one positive tube".

A similar logic is applied to the other mixes of tubes illustrated in BAM.

 

So the basic answer to yr OP is No.

[BALA Prasanna]

Hi Charles,

                  I assume you are referring to generic E.coli. - Yes I am referring Chapter 4 in BAM.

 

I cannot see any procedural option in BAM for E.coli which offers <10 cfu/g. Please clarify. - IN BAM Chapter 4, we can give the result < 10cfu/g by plating method.

 

I presume the <3 refers to the MPN method for a 3x3  setup, ie For 3 tubes each at 0.1, 0.01, and 0.001 g inocula

In fact you can get even more exact by using more tubes. This is detailed in BAM. - Currently we are using 5 tubes (3x5 setup) for each dilution.

[Charles.C]

Hi Charles,

                  I assume you are referring to generic E.coli. - Yes I am referring Chapter 4 in BAM.

 

I cannot see any procedural option in BAM for E.coli which offers <10 cfu/g. Please clarify. - IN BAM Chapter 4, we can give the result < 10cfu/g by plating method.

I am guessing you refer to this -

There are many other methods for enumerating coliforms and E. coli , including several that uses fluorogenic reagents like MUG or other chromogenic substrates for presumptive detection and identification of coliform and E. coli in foods. Many of these tests, such as the Petrifilm dry rehydratable film, the hydrophobic grid membrane filter/MUG (HGMF/MUG) method (13), ColiComplete disc (16), Colilert (AOAC 991.15), have been evaluated by collaborative studies and adopted as official first or final action by the AOAC. There are also many modifications of the membrane filtration assays that have been developed for testing for coliform, fecal coliform and E. coli and some of these may be useful in testing foods such as milk and beverages, but they are used mostly for water, environmental waters, and shellfish harvest waters analysis (5, 7, 20, 22, 23, 31).

or perhaps -
 

Alternatively, E. coli colonies can be distinguished among the coliform colonies on VRBA by adding 100 µg of 4-methyl-umbelliferyl-β-D-glucuronide (MUG) per mL in the VRBA overlay. After incubation, observe for bluish fluorescence around colonies under longwave UV light. (see LST-MUG section II for theory and applicability.)

Note that the MPN method was developed to assist where plate counting was insufficiently sensitive.

 

I presume the <3 refers to the MPN method for a 3x3  setup, ie For 3 tubes each at 0.1, 0.01, and 0.001 g inocula

In fact you can get even more exact by using more tubes. This is detailed in BAM. - Currently we are using 5 tubes (3x5 setup) for each dilution.

Then yr value of 3 should probably be replaced by 1.8

 

 

Conventionally results for E.coli are typically reported numerically, unlike zero-tolerant pathogens such as Salmonella which (ideally) use a "not detected in X gm " style

[BALA Prasanna]

Hi Charles,

                   IN BAM method we can't give the result absent for E.coli.

 

In IS method we can give as Absent/g. 

 

But the concern is, our team was put in the specification Absent by using the method of BAM.

 

Now we are changed the specification. But earlier results need to close it.

 

So that we need to justify the result  <3MPN/g or else <10 cfu/g is same as absent in g

[Hank Major]

<3 is defined as negative. absent, or zero. There's no need to justify it since that is outside the scope of what you are doing. Even if you were a laboratory seeking certification, you would be judged on how well you follow the published protocol, not on any "justification".

[Charles.C]

<3 is defined as negative. absent, or zero. There's no need to justify it since that is outside the scope of what you are doing. Even if you were a laboratory seeking certification, you would be judged on how well you follow the published protocol, not on any "justification".

 

Hi Hank,

 

I sort of agree with yr conclusion but not yr first sentence.

 

The precise solution for the general mathematical formula for predicting the MPN value is, when all MPN tubes give a negative result, unfortunately "indefinite". This is the (simplified) reason for the FDA's "definition" as quoted in Post 3.

 

Some older MPN tables do in fact show "zero" which is user-wise intuitively, engagingly, logical but is nonetheless mathematically "unjustifiable". And similarly for "negative" and "absent".

 

It might be added that the particular disadvantage of MPN data is their "large" confidence intervals. An EU think-tank on this subject suggested that official specifications using MPN data as "M" values in "nmcM" type micro. limits should have a tolerance "uplift" due this characteristic however, afaik. the recommendation has never been implemented.

[Hank Major]

Hi Hank,

 

I sort of agree with yr conclusion but not yr first sentence.

 

The precise solution for the general mathematical formula for predicting the MPN value is, when all MPN tubes give a negative result, unfortunately "indefinite". This is the (simplified) reason for the FDA's "definition" as quoted in Post 3.

 

Some older MPN tables do in fact show "zero" which is user-wise intuitively, engagingly, logical but is nonetheless mathematically "unjustifiable". And similarly for "negative" and "absent".

 

It might be added that the particular disadvantage of MPN data is their "large" confidence intervals. An EU think-tank on this subject suggested that official specifications using MPN data as "M" values in "nmcM" type micro. limits should have a tolerance "uplift" due this characteristic however, afaik. the recommendation has never been implemented.

 

Of course, you are correct that the results are "indefinite".  For example, one cannot take the log of zero.  One might have to use a placeholder value of 1 for one's statistical analyses. 

[BALA Prasanna]

Dear Team,

 

 

Thanks for your valuable feedback.