2 replies to this topic

[John Schrils]

At our company we manufacture nutritional supplements and currently have a specification of <10 CFU/g for Total Coliforms on all raw materials and finished goods. This is a spec that pre-dates my time here but I cannot find the answer to the question: "Where did that limit come from?"

I have examined all relevant chapters concerning the subject of coliforms within the USP, the Code of Federal Regulations (U.S.), the FDA's Bacteriological Analytical Methods, and the AOAC guide. While most of these texts (other than CFR) describe methodologies for testing, none of them state any limits and none of the literature states any limits for coliforms for items other than water (including bottled), shellfish, seafood, raw milk, and at times juice (HACCP).

Coliforms are a "working definition" based upon phenotypic similarities to E. coli, and I believe it is now well understood that there is not a true, direct, undeniable link between total coliform presence and sanitary practice, especially when dealing with products containing botanicals and items of animal origin which bring their own biological load to the table as raw materials. Thus "all E. coli are coliforms but not all coliforms are E. coli", which ultimately is the true organism of concern. My questions therefore are:

1. Is there a regulatory limit on coliforms for foods and supplements that is "on the books" anywhere? Or is it one of those "that's just the way we've always done it" without really knowing why situations. I'm particularly interested in U.S. and Canadian regulation, but also E.U.

2. If so, what is the justification for said levels?

Any input would be greatly appreciated.


Regards,

John Schrils

[Charles.C]

Dear John Schrils,

(first 3 paragraphs in italics added later after more thought )

Food supplements not my area of expertise so I originally answered yr query in a generic way (see normal text below). However since you requested any locatable micro.specs with coliform data, you might be interested in this document –
 AHPA_GuidancePolicies.pdf   168.96KB   750 downloads
The coliform limits mentioned are either higher or much higher than your posted limit depending on the nature of the material. Assuming your original reference to botanicals relates to yr actual interest I suppose the values given may be relevant and are presumably based on experience within the trade. The very forceful caveats before and after the specs regarding precautions required if using the stated limits is daunting but probably very realistic. I think many textbooks suggest that detailed micro.limits in this product area often have limited, or no, value for raw material inputs. Of course, if you have a more controlled situation, a usable baseline may be achievable.

As per my example at end, offhand, yr spec. looks more like a verification of a finished product where processing has reduced items such as coliform or else relate to particularly “clean” raw material. Perhaps you know that already ? It would be very surprising IMO if random botanicals all passed yr coliform limit and, if so, I would probably hasten to validate my coliform procedure (??).

(further added - I also noticed this cautious comment by USP -

Coliform or Enterobacteriaceae counts will not be an appropriate general microbiological criterion for botanicals. However, when it is considered advantageous, Coliform or Enterobacteriaceae counts may be included in the individual monographs

http://www.pharmacopeia.cn/v29240/usp29nf24s0_c2023.html

It's an interesting question but has a wide scope. As you probably already determined, there are considerable geographical variations/opinions with respect to food micro.regs.

I believe you can find limits for coliforms for various other foods (ie not in yr original lis) although the precise interpretation of many limits is often open to debate. You can find discussions on the "basic" (+/-) merits of most of the "proposed" sanitary indicators, eg APC, coliforms, E.coli in standard food micro. textbooks. The specific validity may well depend on features like the process / input material. For example, coliform data is likely to be more readily (local sanitation-wise) interpreted for microbiologically pristine seafood raw material from cold water sources or product from cooking processes where the original vegetative micro-flora is largely eliminated.

It is probably true that the inclusion in governmental regulatory limits is decreasing, eg coliforms not present in EC 2073/2005 micro.regs for foodstuffs but I think you will find many individual countries still add in. (even fecal coliforms is still in use in some places although considered obsolete/defective in many opinions). Maybe just a question of official inertia.

Some examples of the logic driving micro-limits can be seen in the attachments below which are getting a bit old now but, i think, are still regarded as important publications, –

 micro-organisms in foods 2.pdf   802.98KB   821 downloads
 development-use of Microbiological criteria.pdf   2.59MB   676 downloads

The document below (very) critically surveyed available information on micro.limits / utlilised strategies ca. 1999 and acted as a lead-in to the final, detailed EC micro regs. control document (No.2073) of 2005.
As an example comment, pg 16 –

(4) Many of the microbiological criteria do not appear to be meaningful in terms of consumer health protection, for example, aerobic plate counts and coliform counts in certain foods

 evaluation mic.criteria food products 1999.pdf   162.38KB   536 downloads

Re "<10"

One explanation to yr query could come from a coliform measurement via VRBA plates (method probably in BAM) requiring that no colonies be detected on lowest dilution plates. (eg 1ml solution plated contains 0.1g of sample).
Accordingly, 1 colony > 10 cfu/g, so absence is equivalent to <10cfu/g.

It is also popular as “m” in various 3-stage micro. specs (eg coliforms, E.coli) where the exact choice is perhaps due to convenience, (similar to the customary 10-fold link between m and M).
(eg see pg 232 in first attachment)

Rgds / Charles.C

[John Schrils]

Dear Charles,

Thank you so much for your very helpful, thoughtful, and extensive reply.

Regards,

John S.