38 / 258
L
in
et al
.:
J
ournal of
AOAC I
nternational
V
ol
.
100, N
o
.
1, 2017
149
response occurs when biotin is injected at high concentration.
To achieve accurate quantification, the method requires an
appropriate sample size to ensure that the response (peak area)
of the injected sample fits within the established calibration
curve range. Based on the recovery data of Standard
Reference Material (SRM) 1849a with different sample sizes,
the suitable range of the calibration curve is 8.5–60 ng/mL
for this method (Table 2). If the biotin concentration of the
injected sample is >60 ng/mL, a further dilution with mobile
phase is required.
Precision
.
—
All fortified and unfortified matrixes were
freshly prepared and analyzed in duplicate on 6 days. The
SMPRs require RSD
r
to be ≤6% for matrixes that contain >1 μg
biotin/100 g reconstituted final product. The RSD
r
obtained from
the SLV ranged from 0.5 to 3.0% and met the requirements. The
intermediate RSD
R
was in the 1.3–4.5% range (Table 3).
Accuracy
.
—
A total of 11 representative SPIFAN matrixes
were spiked with biotin dissolved in 0.5% ethanol. Fortified
matrixes were spiked at either 100% or 50% of the previously
determined biotin level, and placebos were spiked at either
150% or 50% of the previously determined biotin level in their
corresponding fortified matrixes. The spiked sample was either
stored at room temperature for 2 h or stored refrigerated for 24 h
to allow biotin to become incorporated into the sample matrix.
The spiked samples were prepared and analyzed in duplicate
on 3 days. In addition, SRM 1849a with different sample sizes
was prepared and analyzed. The result was compared to the
certificate of authenticity value.
The SMPRs require recovery to be 90–110% for matrixes
that contain >1 μg biotin/100 g reconstituted final product. The
requirements were met for 10 of 11 tested SPIFAN matrixes. The
mean spike recovery data ranged between 95 and 111% (Table 4).
Recovery of SRM1849a met the requirement (Table 5).
LOQ
.
—
Biotin LOD and LOQ values were determined
experimentally by spiking a very low level of biotin into placebos.
Blank mean and SD were obtained from eight injections.
LOQ = blank mean + 10 SDs
The result was confirmed by the S/N method. A placebo was
spiked with biotin at the estimated LOQ level and the peak
S/N was 10.
The LOQ was estimated to be 0.8 μg/100 g reconstituted
final product for powder, assuming a 4 g sample was diluted to
50 mL; and 1.5 μg/100 g for RTF, assuming a 20 g sample was
diluted to 50 mL. The biotin levels in all the SPIFAN matrixes
are above the LOQ of this method.
Figure 2. Representative calibration curve.
Table 2. Suitable range of sample concentration after dilution
SRM
Product
Code
Reference value,
mg/kg powder
Sample
size, g
HPLC-injected
sample concn,
ng/mL
Measured value,
mg/kg powder
Recovery versus
CoA, %
a
Calibration standard
Suitable range
concn, ng/mL
Level
Concn, ng/mL
1849a
CLC-10b
1.99
0.1445
6.276
2.172
109
Blank
0
8.5–60
0.1532
6.981
2.278
114
1
4.981
0.2006
8.454
2.107
106
2
9.962
1.0020
40.21
2.006
101
3
19.92
1.0006
41.05
2.051
103
4
39.85
1.5177
58.69
1.934
97.2
5
59.77
1.5082
57.60
1.910
96.0
6
79.70
2.0006
69.39
1.734
87.1
7
99.62
2.0003
69.28
1.732
87.0
a
CoA = Certificate of Analysis.
38