145 / 258
Butler-Thompson et al.
: J
ournal of
AOAC I
nternational
V
ol.
98, N
o.
6, 2015
1667
pediatric, and adult nutritionals, and none of the published
myo-inositol methods will only determine both free myo-
inositol and myo-inositol bound as phosphatidylinositol.
In 2011 and 2012, the AOAC Expert Review Panel
(ERP) on SPIFAN Nutrient Methods granted First Action
status to two inositol methods, AOAC
2011.18
(4) and
AOAC
2012.12
(5). Both of these methods use ion
chromatography and pulsed amperometric detection with
a gold electrode, but they differ in their sample preparation
procedures and chromatographic separations. Single-
laboratory validations (SLVs) were completed with both of
these methods, and in March 2013AOAC
2011.18
was selected
by the ERP for further evaluation with a multilaboratory
collaborative study to determine method reproducibility (4, 6).
Consistent with the requirements of the SMPRs, the
method chosen by the ERP for a multilaboratory collaborative
study (AOAC
2011.18
) determines only free and
phosphatidylinositol bound myo-inositol. With this method,
free myo-inositol and phosphatidylinositol bound myo-
inositol are extracted using two different sample preparation
procedures, separated by ion chromatography using a
combination of Dionex PA1 and MA1 columns with column
switching, and detected with pulsed amperometry using a
gold electrode. Free myo-inositol is extracted from samples
with dilute hydrochloric acid and water. Phosphatidylinositol
is extracted from samples with chloroform and separated
from other fats with silica SPE cartridges. Myo-inositol is
then released from the glycerol backbone with concentrated
acetic and hydrochloric acids at 120°C.
Multilaboratory Collaborative Study
Initially 15 laboratories expressed interest in completing the
multilaboratory collaborative study, but only 10 laboratories
were able to participate. The 10 participating laboratories
were located in five different countries. The remaining five
laboratories were not able to participate because of time and
resource constraints and issues with the importation of samples
into their countries. One of the participating laboratories did not
receive the liquid products, and one laboratory was only able to
complete the free myo-inositol testing.
Before actual multilaboratory collaborative study samples
were analyzed, each participating laboratory was asked to
analyze a practice sample, Standard Reference Material (SRM)
1849a, in duplicate in order to identify and resolve any testing
issues that they may have had executing the method. During the
analysis of the practice sample for phosphatidylinositol bound
myo-inositol, it was discovered that there were significant
differences between the ovens used by the study participants.
Even though all oven temperatures were set at 110°C, it appeared
that in some laboratories samples did not receive enough heat
during the hydrolysis step to efficiently release bound myo-
inositol. In all cases when laboratories were asked to increase
oven temperatures 10–20°C, myo-inositol recoveries improved.
After approval of the practice sample results by the Study
Directors, laboratories began testing the study samples.
Each participating laboratory received blind duplicates of nine
SPIFAN matrixes fortified with myo-inositol or with significant
levels of inherent myo-inositol. The SPIFAN matrixes analyzed
included SRM 1849a, an infant formula partially hydrolyzed
milk based powder, an infant formula partially hydrolyzed
soy-based powder, a child formula powder, an infant elemental
powder, an infant formula milk-based powder, an infant formula
soy-based powder, an infant formula milk-based RTF, and an
unfortified infant formula milk-based RTF.
Per SPIFAN requirements, participants were asked to
reconstitute all powders prior to analysis. SRM 1849a 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 of powder in 200 mL laboratory water. For
free myo-inositol analyses, participants were asked to analyze
all 18 samples on one day, and for phosphatidylinositol bound
myo-inositol analyses participants were asked to split the
18 samples into two groups of 10 and eight according to the
data reporting sheets in the study protocol and to test each
group on a separate day. Although the original AOAC First
Action method
2011.18
published in the
Journal of AOAC
INTERNATIONAL
(4) included a step to dilute each sample
with water before adding 50% sodium hydroxide after sample
hydrolysis, this step was accidentally omitted from the Official
First Action method on the AOAC website and the original
AOAC
2011.18
multilaboratory study protocol. This error
was discovered after one of the participating laboratories
reported a more vigorous than expected reaction when 50%
sodium hydroxide was added to the hydrolyzed sample. The
protocol was corrected, and the revised protocol was sent to
all study participants. It should also be noted that the original
method validation data published in the First Action
2011.18
manuscript (4) were generated using direct powder weights.
Upon completion of the sample analyses, participating
laboratories were asked to send all of their data to the Study
Directors. This included all standard and sample chromatograms
for the instrument check, practice and test sample analyses,
standard curve information, calculations, and completed
Reporting of Analysis Forms with dilution and sample weights.
Participants were also asked to report any deviations to the
method and any relevant comments based on their experiences
with the method.
All data were statistically analyzed according to AOAC
INTERNATIONAL guidelines to determine overall mean,
repeatability SD (s
r
), repeatability RSD (RSD
r
), reproducibility
SD (s
R
), reproducibility RSD (RSD
R
), and Horwitz ratio
(HorRat; 7). Cochran (
P
= 0.025, one-tail) and Grubbs’ (single
and double,
P
= 0.025, two-tail) tests were used to determine
statistical outliers.
Myo-inositol SPIFAN SMPRs for repeatability were
≤5% RSD at myo-inositol concentrations of 2–68 mg/100 g
RTF liquid. Myo-inositol SPIFAN SMPRs for reproducibility
were ≤8% RSD in products with myo-inositol concentrations
ranging from 2 to 68 mg/100 g of RTF liquid.
145