548 / 670
Memo
Monique Steegmans
16001n-MSS-MSS-SLV FOS01 AOAC997.08 Beneo-Orafti, 1 Febr 2016, page 10/18
One of the most important remarks for the AOAC 997.08 method is that the accuracy
decreases if the amount of sucrose is very high compared to the fructan content. So the ratio
of sucrose against fructan is given in the table below.
Product
SPIFAN
Code
ID
number
Fructan
g/100g*
Sucrose
g/100g*
Ratio
Sucrose /
Fructan
Adult nutr powder Milk Protein based
11750017V3
J4G-
00019
0,59
1.1
1.8
Adult nutr RTF high Fat
00729RF00
J4G-
00980
0,06
0.3
5
Child formula powder
00411RF00
J4G-
00013
0,35
3.3
9.4
Child formula powder
00412RF00
J4G-
00974
0,28
3.2
11.4
*results are expressed in g/100 g reconstituted final product.
The fructan amounts obtained are not really covering the proposed high values of the range (5%)
however in commercial samples regularly analyzed in our lab, the highest values observed are
around 1,5g fructan/100g reconstituted final product. Further in this report, the determination of the
recovery was done by spiking samples over the total analytical range until 5.0%.
The lower the concentration the higher the RDSr as the error of the method increases.
2 samples (J4G-00019 and J4G-00013) show an RDSr lower than the proposed 6% and 2
samples (J4G-00980 and 974) have a higher RSDr. These higher values could be explained for
one sample (J4G-00980) as a very small amount of fructans (0.06) was added. The variation in the
detailed results (annex4) is very small being 0.05 to 0.06 but because of this low amount the RSD
is quite high.
The other sample J4G-00974 with a RSDr of 9.5% has a similar concentration of fructans and
sucrose as sample J4G-00013 so probably other matrix effects might have an influence.
For this SLV the average RSDr is 6.3 % which meets the requirement of the SMPR 2014.002. If
the low amount of 0.06 is skipped the average RSDr is 5.7%.
Detailed results can be consulted in annex 4.
At the time of the collaborative study of AOAC 997.08, the RSDr ranged from 2.9 to 5.8%, but
concentrations of fructan were higher, 4.6 to 42 g/100g sample.
These samples are part of a Test Materials Kit ordered at AOAC, containing samples fortified with
unknown amounts of fructans so the recovery can only be evaluated by comparing the figures of
these results with the spiked levels.
This part should also be taken into consideration to evaluate the performance of all methods
proposed for SPIFAN-FOS.
Fos-01 (February 2016)
FOR ERP USE ONLY
DO NOT DISTRIBUTE