6. AOACSPIFANMethods-2018Awards

155

Butler-Thompson et al. : J ournal of AOAC I nternational V ol. 98, N o. 6, 2015  1677

(b)  Preparation of standard curve.— For each working standard concentration, average the peak areas or heights from each of two consecutive sets of standards. Prepare a standard curve by performing linear least squares regression on the concentrations versus the averaged peak areas or heights. A standard curve must have a correlation of at least 0.999 to be considered acceptable for sample calculations. At each working standard level, the peak areas or heights of standards injected before and after a set of samples must not increase or decrease by more than 7%. (c)  Calculationofmyo-inositolinsamples. —Theconcentration of myo-inositol in a prepared sample is extrapolated from the standard curve prepared above. From the diluted, prepared sample concentration, the product concentration can be calculated: C p = (C d × D 1 )/S where C p is the concentration of myo-inositol in the product sample in mg/kg, C d is the concentration of myo-inositol in the prepared sample in mg/L, D 1 is the dilution volume in mL, and S is the sample weight in g. Note : For each set of samples, the control result must be within 3 SDs of the control mean. All of the free, phosphatidylinositol bound, and free plus phosphatidylinositol bound myo-inositol collaborative study data are summarized in Tables 1–9. It should be noted that laboratory 5 did not receive the liquid RTF study samples and that laboratory 6 was not able to complete the phosphatidylinositol bound myo-inositol testing. Using the AOAC INTERNATIONAL Interlaboratory Study Blind (Unpaired) Replicates Workbook (7), statistical outliers from one or two laboratories were identified in five of the nine products analyzed for free plus phosphatidylinositol bound myo- inositol content. After removal of outliers, RSD r ranged from 0.51 to 3.22% and met the SMPR of ≤5% for all the products analyzed. The RSD R ranged from 2.83 to 7.55% and met the SMPR of ≤8% for all the products analyzed. When the outliers were included, RSD r ranged from 0.91 to 4.67%, meeting the SMPR of ≤5% for all the products analyzed, and the RSD R ranged from 2.83 to 11.2%, meeting the SMPR of ≤8% for eight of the nine products analyzed. Since it is possible that some laboratories may only use this method for free myo-inositol analyses, a review of the free myo-inositol collaborative data is also included here. Using the AOAC INTERNATIONAL Interlaboratory Study Blind (Unpaired) Replicates Workbook, statistical outliers from one or two laboratories were identified in five of the nine products analyzed for free myo-inositol content. After removal of free myo-inositol outliers, RSD r ranged from 0.46 to 3.03% and met the SMPR of ≤5% for all the products analyzed. The RSD R ranged from 2.15 to 12.6% and met the SMPR of ≤8% for eight of the nine products analyzed. When the outliers were included in the free myo-inositol data summary, RSD r ranged from 0.97 to 4.45%, still meeting the SMPR of ≤5% for all the Results and Discussion

products analyzed, and the RSD R ranged from 2.77 to 12.6% and met the SMPR of ≤8% for seven of the nine products analyzed. Several laboratories provided comments about the method. Some laboratories made positive comments regarding the column switching format because it saved time and kept the electrode clean, while other laboratories would prefer using a gradient rather than column switching. One laboratory questioned the need for determining the phosphatidylinositol component. As noted previously, some laboratories had to use a higher temperature than that listed in the method to improve recoveries of phosphatidylinositol boundmyo-inositol, and some laboratories had trouble adding 50% sodium hydroxide directly to the samples immediately after hydrolysis. One laboratory recommended adding additional guidance for determining the amount of acid needed to adjust the pH of a sample to 4.5 for free myo-inositol analyses and for determining the percentage recovery of phosphatidylinositol from the SPE cartridges. One laboratory noted that the method should specify that the hydrolysis procedure be performed in a fume hood. Additional information was added to AOAC Final Action Method 2011.18 to address some of study participants’ comments and concerns listed above. AOAC Method 2011.18 was collaboratively studied by nine to 10 laboratories from five different countries with a variety of infant and pediatric nutritional matrixes. The method demonstrated acceptable repeatability and reproducibility and met the SPIFAN SMPRs for free and free plus phosphatidylinositol bound myo-inositol in most of the matrixes analyzed. Conclusions

Recommendation

Themultilaboratorycollaborative studydatawere summarized and presented to the AOAC ERP in September 2014. After reviewing the data, the ERP voted to move AOAC 2011.18 to Final Action status, and the method was approved by the AOAC Official Methods Board as a Final Action Method.

Acknowledgments

The authors would like to thank the following collaborators and their associates: Sumalee Purachaka, Mead Johnson Nutritionals, Don Huaroh, Muang, Chonburi, Thailand Lillian Chen and Lipika Basumallick, Thermofisher Scientific, Sunnyvale, CA Wu Bolong, Test Center of Chinese Academy of Inspection and Quarantine, Beijing, China George Lautenschlager and Scott Christiansen, Perrigo Nutritionals, Georgia, VT David Ellingson and John Austad, Covance Laboratories, Madison, WI