PMFLIST Archives

May 2006


Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Scott Sutton <[log in to unmask]>
Reply To:
The Pharmaceutical Microbiology Forum Email List <[log in to unmask]>
Tue, 30 May 2006 07:42:15 -0400
text/plain (124 lines)

The question, as I see it, isn't so much whether or not it sounds
reasonable.  "Reasonableness" is frequently based on our desires and
preconceptions.  The question is rather is it scientifically supportable.

Let's look at the question fresh - we can agree that in normal situations
the valid number of bacterial colonies to count on a plate is 25-250 (my
apologies, I was a little sloppy with this range on my previous Email).  We
can agree that if there are more than 250 colonies on a plate, we report the
data out as TNTC (Too Numerous To Count).  So we agree on a countable range
and we agree on how to report excess at the high end.

How do we report excess at the low end of the countable range?  Do we ignore
it?  I suggest we change "accepted practice" and opt for good science in
this regard.  As we introduce more and more alternate, rapid microbiological
methods, we are going to be more often faced with the questions of accuracy
and precision in our test results.  

ISO 17025 "General Requirements for the Competence of Testing and
Calibration Laboratories" is an interesting read in this regard.  In section of "Validation of Methods" you can find the statement "The
techniques used for the determination of the performance of a method should
be one of, or a combination of, the following: . . . assessment of the
uncertainty of the results based on scientific understanding of the
theoretical principles of the method and practical experience."  This
statement is only one of several dealing with validation (accuracy,
precision, range, etc - see USP <1223> or Pharm Eur 5.1.6) and uncertainty
of measurements to be found in that ISO standard.  

A question for testing labs certified under ISO 17025 - How do you meet the
requirements for understanding and reporting uncertainty in the data?

Back to the question at hand.   How do we report out data from a plate that
has too few colonies to be in the countable range?  Common practice is to
ignore the situation and report it out as if it were an accurate count.  I
have even seen situations were a lab will take duplicate plates with counts
of 0 and 1 and report out 0.5 CFU at a 1/10 dilution = 5 CFU/mL.  This
practice is not scientifically supportable.

Merely because it would be convenient to report out CFU down to 1 CFU/plate
is not a justification for doing so.  These numbers are not even remotely
accurate and we are only digging deeper holes for ourselves by participating
in the charade - holes that will come to present major obstacles when we
attempt to validate an alternate method and are FORCED to apply rigorous
measures under USP or Pharm Eur guidance.

Yes, in answer to your question, the only defensible act in this situation
(in my opinion) is to acknowledge that the plate count method has a Limit of
Quantification, and the LOQ is 25 CFU/plate.  Pursuing this argument, if you
have counts that fall below the countable range of 25-250 CFU/plate on the
1/10 dilution you would report the data out as <250 CFU/mL.  Even if the
duplicate plate counts are 24 and 23 - they fall below the LOQ of the plate
count method.


Scott Sutton, Ph.D.
Pharma Consultant, Microbiology
Vectech Pharmaceutical Consultants
[log in to unmask]
This Message is privileged, confidential and protected by law from
disclosure. If you received this message in error, then forward it to
[log in to unmask] and delete it from your system.

-----Original Message-----
From: The Pharmaceutical Microbiology Forum Email List
[mailto:[log in to unmask]] On Behalf Of Linda K. Skowronsky
Sent: Friday, May 26, 2006 5:57 PM
To: [log in to unmask]
Subject: Re: [PMFLIST] Bioburden

In reply to Scott's response....I'm confused. Based on what you said, then 
if we have no growth on 2 plates from a 1:10 dilution and we can only 
report cfu in the countable range of 20-200, does this mean the minimum 
detection limit would be <200? Does this sound reasonable??? Or perhaps I 

Best Regards,

The PMFList ( is operated from
The Microbiology Network ( and supported by
our sponsors ( as a service to
the scientific community.

Please take a second to visit our sponsors' web sites and say thank you for their support of this service.

Accugenix -

American Type Culture Collection -

ATS Laboratories -

Biolog -

Biomerieux -

Dupont Qualicon -

EMD Chemicals -

EMSL Analytical, Inc. -

Genomic Profiling Systems, Inc. -

MIDI, Inc. -

Millipore -

Pall -

NovaTek International -

Raven Biological Labs -

Veltek Associates, Inc -

The nature of this service is to provide a medium for communication.  The specific statements and endorsements of individuals participating in the discussions are not necessarily those of The Microbiology Network, Inc., the PMF, or the sponsors of the list.