111 / 363
Bidlack et al.:
J
ournal of
AOAC I
nternational
V
ol.
98, N
o.
5, 2015
1383
RP chromatography, and it has been shown to under-recover
vitamin K in some more complex infant, pediatric, and adult
nutritional matrixes (5, 6).
This new NP HPLC method with postcolumn reduction and
fluorescence detection allows for the quantitative determination
of
trans
vitamin K
1
in infant, pediatric, and adult nutritionals.
Vitamin K
1
is extracted from products with iso-octane after
precipitation of proteins and release of lipids with methanol.
Prepared samples are injected onto a silica HPLC column where
cis
and
trans
vitamin K
1
are separated with an iso-octane–
isopropanol mobile phase. The column eluent is mixed with a
dilute ethanolic solution of zinc chloride, sodium acetate, and
acetic acid, and vitamin K
1
is reduced to a fluorescent derivative
in a zinc reactor column. The resulting hydroquinone is then
detected by fluorescence at an excitation wavelength of 245 nm
and an emission wavelength of 440 nm.
Single-Laboratory Validation (SLV)
Experimental
To verify the applicability of this method, an SLV with all 12
SPIFAN infant, pediatric, and adult matrixes was completed.
To establish method precision, all fortified and unfortified
matrixes were prepared and analyzed in duplicate on 6 days.
National Institute of Standards and Technology Standard
Reference Material (SRM) 1849a (Infant/Adult Nutritional
Formula) was reconstituted by dissolving the entire contents
of the sachet (10 g) in 90 mL water. All other powders were
reconstituted by dissolving 25 g powder in 200 mL laboratory
water. New reconstitutions were prepared each day.
Method accuracy was established by spiking aliquots of
each SPIFAN matrix with vitamin K
1
at approximately 50 or
100% of the previously determined
trans
vitamin K
1
level.
Approximately 40–250 µL vitamin K
1
, dissolved in ethanol,
was added to 25 g aliquots of each SPIFAN matrix, and all
sample and spike weights were recorded. On 2 days each
sample matrix was spiked at 100% of the previously determined
trans
vitamin K
1
level, and on the third day each matrix was
spiked at 50%. On each day spike blanks were also prepared by
adding 75–250 µL vitamin K
1
, dissolved in ethanol, to 25 mL
iso-octane. Spiked samples and blanks were thoroughly mixed
and stored refrigerated for at least 24 h to allow vitamin K
1
incorporation into the sample matrix. After at least 24 h, spiked
and unspiked samples were prepared and analyzed in duplicate
as described in the method. Spike blanks were diluted to
appropriate concentrations with iso-octane and analyzed along
with the spiked and unspiked sample preparations.
Method linearity was evaluated by injecting five or six
standards with
trans
vitamin K
1
concentrations ranging from
approximately 2 to 90 µg/L before and after every set of
samples analyzed during validation. Calibration curves were
Table 1. Method performance requirements:
trans
-vitamin K
1
a
Analytical range
1–100
b
LOQ
≤1
b
RSD
r
, %
1–10
b
≤8
>10
b
≤5
Recovery, %
90 to 110 of mean spiked recovery
over the range of the assay
RSD
R
, %
1–10
b
≤15
>10
b
≤10
a
Concentrations apply to (a) RTF liquids “as is”, (b) reconstituted
powders (25 g into 200 g of water), and (c) liquid concentrates diluted
1:1 by weight.
b
µg/100 g reconstituted final product.
Table 2.
Trans
vitamin K
1
SLV data—precision
Sample type
No. of replicates
(duplicates on multiple days)
Mean,
µg/100 g RTF SD
r
RSD
r
, % SD
IP
RSD
IP
, %
Child formula powder, placebo
6
2.03
0.040
2.0
0.045
2.2
Infant elemental powder, placebo
6
2.03
0.020
1.0
0.025
1.2
Adult nutritional RTF high protein, placebo
6
3.47
0.021
0.6
0.040
1.2
Adult nutritional RTF high fat, placebo
6
3.07
0.034
1.1
0.034
1.1
Infant formula RTF milk based, placebo
6
2.16
0.038
1.8
0.038
1.8
SRM 1849a
12
1.11
a
0.022
2.0
0.025
2.3
Adult nutritional powder milk protein based
12
3.26
0.097
3.0
0.142
4.4
Infant formula powder partially hydrolyzed milk based
12
7.69
0.123
1.6
0.170
2.2
Infant formula powder partially hydrolyzed soy based
12
8.99
0.141
1.6
0.226
2.5
Adult nutritional powder low fat
12
2.92
0.102
3.5
0.102
3.5
Child formula powder
12
2.66
0.049
1.8
0.050
1.9
Infant elemental powder
12
7.57
0.129
1.7
0.451
6.0
Infant formula powder milk based
12
6.09
0.105
1.7
0.131
2.2
Infant formula powder soy based
12
6.26
0.105
1.7
0.211
3.4
Infant formula RTF milk based
12
9.01
0.168
1.9
0.189
2.1
Adult nutritional RTF high protein
16
9.10
0.283
3.1
0.299
3.3
Adult nutritional RTF high fat
12
10.7
0.120
1.1
0.179
1.7
a
Results reported as mg/kg powder.
Candidates for 2016 Method of the Year
110