690 / 1195
VIDAS LPT - Summary 2012
EUROFINS IPL Nord SAS
–
1 rue du Professeur Calmette
–
59046 LILLE cedex
p 10/14
3.3.3 Comments (discrepancies with respect to expected results, exclusions, etc.)
The uncontaminated samples (level L0) were all found to be negative with both methods.
For the lowest level of contamination (level L1), of the 112 samples tested, 72 were found to be positive with the
alternative method and 64 with the reference method.
Of the 112 samples tested with the highest levels of contamination (level L2), 110 were found to be positive with
the alternative method and 108 with the reference method.
Since the enrichment broth is different between the two methods and the TVC is significant (up to10
8
CFU/g), the
most likely hypothesis to explain these results is that the growth of the strain was inhibited by the background flora,
and did not reach the level of detection of each of the methods.
Conclusion
Finally, it is thus possible to interpret the results from 14 laboratories (receipt on D1, temperature between 0 and
8°C and exclusion of abnormal results).
The results of the reference method and the alternative method are consistent for the 14 laboratories selected, with
the exception of 12 false-negative samples and 22 additional positive samples for the alternative method.
3.4 Calculation
3.4.1 Specificity (%SP) and sensitivity (%SE) for both methods
For level L0
, it is required to calculate the specificity percentage (%SP) for each of the methods:
SP = {1-(FP/N-)} x 100
where FP, number of false positives
N-, total number of L0 tests
For levels L1 and L2
, it is required to calculate the sensitivity percentage (%SE) for each of the methods, with
respect to the number of expected positive results:
SE = (TP/N+) x 100
where TP, number of true positives
N+, total number of L1 or L2 tests
Level
Reference method
Alternative method
SP/SE
LCL* %
SP/SE
LCL* %
L0
SP% =
100,0
100
SP% =
100,0
100
L1
SE% =
57,1
48
SE% =
64,3
55
L2
SE% =
96,4
93
SE% =
98,2
96
L1+L2
SE% =
76,8
69
SE% =
81,3
74
* LCL: low critical value, defined by the EN ISO 16140 standard
3.4.2 Relative accuracy (AC)
The relative accuracy is calculated using the following formula:
AC = {(PA + NA) / N} x 100
where PA, number of positive agreements
NA, number of negative agreements
The results of the pairs of results from the alternative method and the reference method are given below.
Positive reference method
(R+)
Negative reference method
(R-)
Positive alternative method
(A+)
Positive agreement (A+/R+)
PA = 160
Positive deviation (R-/A+)
PD = 22
Negative alternative method
(A-)
Negative deviation (A-/R+)
ND = 12
(1)
Negative agreement (A-/R-)
NA =142
(1)
not including any non-confirmed positive result with VIDAS LPT