"Loh Han Liat" <hanliat from tp.edu.sg> wrote in message
news:mailman.643.1225999747.29717.microbio from net.bio.net...
Hi All,
I've got questions regarding the reporting of enumeration results based
on plating. The same issue occurs with several testing methods from
FDA-BAM as well as AOAC, and I really need some good advice.
I'll use the example of enumerating for S. aureus, according to FDA-BAM.
Accordingly, I'll divide 1 ml of inoculum over 3 Baird-Parker agar
plates. After incubation, I'll select plates that have 20-200 colonies,
add up the S. aureus colonies on the 3 plates and report as count per g
X the dilution factor. Of course prior to this, I would have determined
the fraction of "true" S. aureus among the presumptive colonies and all
that.
Now, the problem is I never get anywhere close to 20 colonies on any of
my plates, for example, as in the table below.
Dilution
Plate 1 (0.4 ml)
Plate 2 (0.3 ml)
Plate 3 (0.3 ml)
10-1
5 colonies
4 colonies
5 colonies
Assume that I've already confirmed all 14 colonies are S. aureus, how do
I report my results?
There's a clause somewhere in the manual that says if the count from the
lowest dilution is < 20, I could still use the plates. Does this mean I
could report the results as 140 cfu/g? If not, how should I report the
results?
Also how should I report the results if there're no colonies on any of
the plates?
Any advice, opinions or comments from anyone with any experience in this
area would be greatly appreciated.
Thank you very much!
Regards
HL
Hi Hl
If you are following your protocol exactly then you are no doubt returning
the correct result. The problem with microbiological assays of this type is
that at low levels of isolation there is usually a high degree of "
Uncertainty" associated with the result of obtained At high levels of
uncertainty you really should report your result with a caveat expresing
the range for upper an lower limits within which results might reside.
YOu need to carry out a validation exercise to calculate the degree of
uncertainty associated with this assay in your lab and possibley with
respect to concensus results in ring trials. Uncertainty is a statiscally
quanity related to confidence limits ( therefore functions of varience and
SD) and which takes into account repeatability, reporducability , precision
and bias. There are no set rules for the calculation of uncertainty but
some stable guide lines do exist based on sound statiscal criteria.
Thsi link might help as will many other ; google "Microbiological
uncertainty of meadsurement
www.ifcc.org/ejifcc/vol13no4/130401006.htm
Until then report your results :- Any isolation of confirm SA is
significant ..usually.
Hope this helps
