129 / 159
FOR EXPERT REVIEW PANEL USE ONLY
NOT FOR DISTRIBUTION
2
General comments about the method:
- Method is an improved version of the method of Cuany,
et.al- Improvements include the addition of a method for determining type of fructan
- Reduction of sample handling steps
- Introduction of internal standard
- Reduction of instrument time
- Overall I think the method should be OK
- Only concern would be the difficulty to assign the appropriate commodity factor in samples with a
complex background containing other OS such as maltodextrins, GOS, resistant dextrin etc…
- Sample preparation has not strictly followed SPIFAN guidelines.
Method Clarity:
- The method is clear, but should be re-written in the appropriate AOAC format before going to first
action.
-
Pros/Strengths:
•
Avoids issues with sucrose interference
•
Step to identify fructan and thus the commodity factor
Cons/Weaknesses
•
Two step process requiring identification of fructan ingredient as a first step.
•
Requires a different chromatographic set-up for the 2 different procedures (could part 1 also be ran
with the borate trap?)
Supporting Data
General Comments:
•
Precision data has been collected on 6 SPIFAN samples (new kit?) and 4 of Abbots own products. In
addition 1 product has been analysed in 3 different labs. All precision data looks good.
(RSD (r) below
4% in all cases
which meets SMPR of ≤ 6%)
•
The analysis of spiked samples has been devised in such a way as to cover many matrices. It is likely the
have used several fructan ingredients – but it is not clear (except in the case of the Abbott matrix).
Spiking levels are also not clearly defined (i.e. what is the actual spike level of the 50% and 150%
overspikes – especially in blank matrices). However reported
recoveries are : 92.9 – 108% (within the
90-110% specified in SMPR)
•
It seems that the whole range 0.03 – 5.0 g/100g defined in the SMPR has not been covered. Maximum
level measured is 3 g/100g.
•
LoQ of 0.03 g/100g has been established by spiking fructan in to a concentrated sucrose solution.
Might have been better to also demonstrate in a sample. However one SPIFAN kit sample has a
concentration of 0.035 g/100g and the performance of the method at this level seems to be
acceptable. .
FOS-04 Review Forms