92 / 471
Participants were instructed to record the temperature of this portion upon receipt of the
1
shipment, document results on the Sample Receipt Confirmation form provided and fax to the
2
study director.
3
4
5
Test Portion Analysis
6
7
Collaborators followed the appropriate preparation and analysis protocol according to the
8
method specified for each matrix. For both matrices, each collaborator received 8 test portions
9
(2 high, 2 medium, 2 low and 2 uninoculated replicates). For the analysis of the each matrix by
10
the 3M Petrifilm RYM Count Plate, a 25 g test portion was diluted with 225 mL of 0.1% Peptone
11
Water and homogenized for 2 minutes. Ten-fold serial dilutions of each sample were prepared
12
and a 1.0 mL aliquot of each dilution was plated onto 4 separate 3M Petrifilm RYM Count
13
Plates. For each dilution, 2 of the plates were incubated at 25 ± 1
o
C and 2 of the plates were
14
incubated at 28 ± 1
o
C. Plates were removed from incubation after 48 ± 2 hrs and typical yeast
15
and mold colonies in the countable range (15-150) were enumerated using a standard colony
16
counter. Plates containing greater than 150 colonies were either estimated or recorded as to
17
numerous to count (TNTC). All plates were re-incubated at the appropriate temperatures for an
18
additional 10-14 hours (to reach a total incubation time of 60 hours) and colonies were
19
enumerated a second time in the same manner as the 48 hour time point.
20
Both matrices analyzed by the 3M Petrifilm RYM Count Plate were also analyzed using the
21
BAM and ISO reference method in a paired study design. Serial dilutions for each sample were
22
spread plated in triplicate onto Dichloran-rose Bengal Chloramphenicol (DRBC) agar (raw
23
frozen ground beef patties) or Dichloran 18% Glycerol (DG18) agar (raw almonds). Agar plates
24
were incubated for 5 days at 25 ±1
o
C and typical colonies in the countable range (10-150) were
25
enumerated using a standard colony counter. Plates containing no colonies were re-incubated for
26
2 days and observed for typical growth.
27
28
29
30
Statistical Analysis
31
32
Each collaborating laboratory recorded the CFU/g results for the reference methods and the
33
3M Petrifilm RYM Count Plate on the electronic spreadsheet provided in the collaborator study
34
outline. The data sheets were submitted to the study director at the end of each week of testing
35
for analysis. The data from each duplicate plates (3M Petrifilm RYM Count Plate) or triplicate
36
plates (BAM and ISO) were averaged. The averaged counts were converted into logarithms for
37
data analysis. Outliers were identified using the Cochran and Grubbs’ tests. A paired t-test was
38
conducted to determine if the mean of replicate samples at each contamination level for each
39
matrix was different between the 3M Petrifilm RYM Count Plate and the reference methods.
40
[3]. A p-value greater than the standard alpha value of 0.05 indicated no statistical difference
41
between the two methods. The repeatability (s
r
), reproducibility (s
R
), relative standard deviation
42
of repeatability (RSD
r
) and relative standard deviation of reproducibility (RSD
R
) of the 3M
43
Petrifilm RYM Count Plate and reference methods were determined by the mean of the
44
logarithm transformations of the counts for each contamination level of each matrix [7]. A lower
45
standard deviation value indicated a greater propensity for repeatability of a method. In addition,
46
the difference of means and reverse transformed difference of mean for each contamination level
47
was determined. [3].
48
49
Candidates for 2016 Method of the Year
78