778 / 1195
For the high level inoculum, dLPOD
C
values of 0.00 (-0.03, 0.03) were obtained between the
1
AOAC OMA 993.12 method and the VIDAS LMX method. The confidence intervals obtained
2
for dLPOD
C
indicated no significant difference between the two methods. dLPOD
CP
values of
3
0.00 (-0.03, 0.03) were obtained between presumptive and confirmed VIDAS LMX results. The
4
confidence intervals obtained for dLPOD
CP
indicated no significant difference between the
5
presumptive and confirmed results. Detailed results of the POD statistical analysis are presented
6
in Table 2013.1B below, and Tables 2013.2C-2D and Figures 2A-2D of Appendix 1.
7
8
ALOA Chromogenic Agar
9
10
Confirmatory results obtained from the ALOA chromogenic agar were identical to the results
11
obtained from the OXA agar specified by the AOAC OMA 993.12 method. Out of a total of 447
12
positives detected by the VIDAS LMX, 445 were confirmed positive using OXA or ALOA
13
selective agars.
14
15
Discussion
16
17
No negative feedback was reported to the study directors from the collaborating laboratories in
18
regards to the performance of the VIDAS LMX assay or the ALOA chromogenic agar. Many
19
laboratories indicated difficulty in identifying and isolating colonies from samples when using
20
the OXA plates, but not from test portions analyzed by the VIDAS LMX method. This may be
21
due to the higher selectivity of the ALOA agar to isolate and differentiate typical
Listeria
22
colonies from competing microflora, such as
Bacillus
colonies. The high selectivity of the
23
proprietary broth, the high background flora and the low selectivity of the OXA agar most likely
24
contributed to this observation as well.
25
For the analysis of 25 g test portions by the VIDAS LMX method, 2 false positives were
26
obtained. The test results produced by the false positive test portions (average test value of 0.22)
27
were much lower than the test values observed with true positives (average value >2.00). By the
28
time the coordinating laboratory received the results, the primary enrichments for these samples
29
had been discarded so no subsequent analysis on the VIDAS LMX was possible. However, the
30
agar plates for these test portions were shipped to the coordinating laboratory for further
31
analysis. Up to 20 different colonies were picked for morphological and biochemical analysis
32
using VITEK 2 GP and no
Listeria
colonies were identified. Additionally, the entire lawn of
33
growth from each agar plate was swabbed, enriched in LMX broth and incubated for 26-30 hours
34
at 37 ± 1
o
C. An aliquot from each tube was analyzed by the VIDAS LMX assay and negative
35
results for
Listeria monocytogenes
were obtained. Results of this investigation lead the study
36
directors to believe that the false positives were the result of cross-contamination during the
37
analysis of the samples.
38
For the analysis of both the 25g and 125g test portions, Laboratory 11 detected the presence of
39
multiple types of
Listeria
spp. An investigation into the results indicated that colonies picked for
40
confirmation did not meet the characteristics of
Listeria
spp., i.e. colonies produced Gram
41
negative or Gram positive with sporesstain reactions and were negative for motility and catalase.
42
The results of these tests should have precluded analysis using the API strips which lead to an
43
inaccurate identification. Due to the fact that final results reported were inconsistent with
44
biochemical results, data produced by Laboratory 11 was removed from the statistical analysis of
45
both the 25g and 125g test portions.
46
Using the POD statistical model, no significant differencein the number of positive results
47
obtained between the two methods being comparedwas observed at both the low and high
48
12