962 / 1195
Eurofins IPL Nord SAS – 1 rue du professeur Calmette – 59046 LILLE cedex
7 / 15
VIDAS LMX – Summary 2012 v01
2.1.5 Analysis of discrepant results
According to annex F of the EN ISO 16140 standard, the minimum number of discordances for which a statistical
test must be conducted in order to compare the two methods is 6.
In this study, among 384 analyzed samples and 190 positive results, 37 results were discordant between both
methods : 18 false negative results and 19 additional positive results. A statistic test was performed.
As the number of discordances was greater than 22, the McNemar test with the
χ
2
distribution for one degree of
freedom was used.
The aim is the determination of d =
PD – ND
and the comparison between d and a d value, depending on the
total number of discordances and according to the EN ISO 16140 (appendix F). Both methods would be considered
as different if d > dvalue.
Number of discordances
d minimal
d
Conclusion
37
12
19 – 18
= 1
Equivalence
The performances of the VIDAS LMX method can be considered as
equivalent
to the reference method (EN ISO
11290-1/A1) for the detection of
Listeria monocytogenes
because d < d minimal (Annex F).
2.1.6 Comments on confirmations and enrichment broth cold storage before VIDAS test
Comments on confirmations
The positive samples at the conclusion of the VIDAS LMX test were confirmed after isolation from the not heated
enrichments - LMX enrichment for general protocol or LX enrichment for specific protocol) and on two
chromogenic agar : Chrom' ID Ottaviani Agosti (Ottaviani Agosti Agar - bioMérieux) or RAPID'
L.mono
( BIO-RAD).
A single sample (C7), was confirmed only on RAPID'
L.mono
.
Comments on VIDAS LMX tests realized after conservation at 2-8°C for 72 h
Globally, the results obtained after 72 hours of preservation of broth LMX in 2 - 8°C, were identical to those
obtained directly after incubation, with the exception of seven samples
(C7, C12, D16, N1, C21, R3 and X1).
2.2. Relative detection level
The objective was to determine the level of contamination for which less than 50 % of the responses obtained are
positive and that for which more than 50 % of the responses obtained are positives.
Different « food matrix strain » couples were studied in parallel with the reference method and the VIDAS LMX
method, for the six representative studied categories.
The artificial contaminations were realized according to EN ISO 16140 and AFNOR validation rules.
The levels of detection, calculated according to the Spearman – Kärber
(1)
method (LOD50), obtained for each
combination “matrix-strain” are the following:
Relative detection level(CFU/ 25 g or 25 ml)
with confidence interval
(2)
LOD
50
Matrix
Strain
Reference method
Alternative method
Rillette
L.monocytogenes
1/2b
0.7 [0.4 – 1.3]
1.1 [0.6 – 1.8]
Raw milk
L.monocytogenes
1/2b
0.5 [0.3 – 0.9]
0.4 [0.2 – 0.7]
Raw milk cheese
(specific protocol)
L.monocytogenes
1/2b
0.7 [0.4 – 1.3]
0.6 [0.4 – 1.0]
Worn red cabbage
L.monocytogenes
4b
0.5 [0.3 – 0.8]
0.6 [0.4 – 1.0]
Smoked salmon
L.monocytogenes
1/2a
0.5 [0.2 – 0.9]
0.4 [0.2 – 0.7]
Process water
L.monocytogenes
1/2c
0.4 [0.2 – 0.7]
0.3 [0.2 – 0.6]
(1)
”Hitchins A. Proposed Use of a 50% Limit of Detection Value in Defining Uncertainty Limits in the Validation of Presence-Absence
Microbial Detection Methods, Draft 10
th
December, 2003.”
(2)
”LOD
50
: estimation of level of contamination enabling positive detection by alternative method in 50 % of cases.
The level of detection of the VIDAS LMX method was between 0.2 and 1.8 CFU/25 g. That of the reference
method was between 0.2 and 1.3 CFU/ 25 g.