Single Lab Validation Report for GOS in Infant Formula and Adult Nutritionals
5
2.2
Method
The method description is added as appendix 5.1.
2.3
Validation Design
2.3.1
Linearity
Galactose and lactose are used to calibrate the system. Three different galactose and lactose stock
solutions were prepared. These were diluted to prepare three series of standard curves spanning the
concentration range from 0.5 to 20 µg/mL.
Arabinose was used as internal standard. The ratio of the peak areas galactose/arabinose and
lactose/arabinose were plotted against the concentration of respectively galactose and lactose (since
the arabinose concentration remained constant throughout).
A linear and quadratic regression was performed, and the residues were calculated and plotted against
the corresponding galactose or lactose concentration.
2.3.2
Precision
All samples of the SPIFAN kit were analyzed once to determine which formulas contain GOS. Three of
the samples from the SPIFAN kit appeared to contain GOS: sample 8 IF Powder Partially Hydrolyzed Milk
Based (1172572116), sample 13 Infant Formula Powder Milk Based (D04HTCVV) and sample 15 Infant
Formula RTF Milk Based (SPIFAN Control Milk Form). Each sample was analyzed in duplicate on 6 days
on 2 instruments. Repeatability, SD(r), and reproducibility, SD(R), were calculated from the data
obtained using the equations below:
Repeatability:
)1 (
)
(
)(
1 1
2
−
−
=
∑∑
= =
kr
x x
r SD
r
i
k
j
i
ij
ij
x
= Individual result on day i (going from 1 to r) within the set of replicates (going from 1 to k)
i
x
= Mean of replicates on day i
r = Number of days
k = Number of replicates
Reproducibility:
1
)
(
) (
1
2
−
−
=
∑
=
n
x x
R SD
n
i
i
i
x
= Individual result within the set of single determinations (going from 1 to n)
x
= Grand mean
2.3.3
Accuracy
Four matrices were selected that covered both infant formula and adult nutritionals: sample 7 (Adult
Nutritional Powder Milk Protein Based), sample 10 (Adult Nutritional Powder Low Fat), sample 11 (Child
GOS-02 SLV
FOR ERP USE ONLY
DO NOT DISTRIBUTE