102 / 112
Exchange ofthimerosal
Page 10
with the mean CV of control samples from a reference Gliadin competitive lot. Several
measurements of similar control samples over the shelf life of this Gliadin competitive lot were used
lot calculate the mean CV of all these measurements for all control samples. The calculated mean CV
was 20.8% (the original data for this reference lot are shown in table 5 in annex section 7.4.2). The
collaborative study for AOAC OMA 2015.05 with the Gliadin competitive had a mean CV of 31.2%
(mean CV of all samples of the collaborative study with a mean concentration above the limit of
quantitation of the Gliadin competitive). The mean CV of the tested Gliadin competitive lot lay with
28.2% between the in house data of the reference Gliadin competitive lot and the mean CV of the
collaborative study. It wasthus considered not to be unusually high and the obtained concentrations
for the substitution of thimerosal with bronidox L were considered to be of value for discussion.
3.4.3
Discussion
In consequence of the higher measurement uncertainty as mentioned in the previous section, the
calculated mean concentrations were less similar to each other than it was observed for the other
three ELISA (see tables2, 4, 6 and 8). Nevertheless, no systematic difference in the concentrations for
the control samples in regard to the used washing buffer could be observed. This demonstrates that
the change of washing buffer does not have a significant influence on the concentrations of the
control samples and thus on the functionality of the test, nor on the stability of the washing buffer.