6. AOACSPIFANMethods-2018Awards

22

H aselberger & J acobs : J ournal of AOAC I nternational V ol . 99, N o . 6, 2016  1577 Figure 1. General formulae for the two major inulin type fructans relevant to SMPR 2014.02 GF � Type: (� − DFru� − (2 → 1)) � − � − D − Fru� − (2 → 1) − � − D − Glc� F � Type: (� − DFru� − (2 → 1)) � − (2 → 1) - D - Fru f 2 H aselberger & Jacobs : J ournal of AOAC I nternati nal Vol. 99, No. 6, 2016

Fru f = fructofuranose Glc p = glucopyranose

Figure 1. General formulae for the two major inulin-type fructans relevant to SMPR 2014.002.

+ 0.1 �� ��� -1

� W

� G

= � 0.9�� avg

�

in the post hydrolysis colorimetric determination of fructan- derived monosaccharides. As a result, the total fructan content will be underestimated. AOAC 997.08 relies on a correction for the monosaccharides released by other carbohydrates, including sucrose (6). This results in compromised precision because of the error propagation in the background corrections which generally limits the usefulness of method 997.08 , or similar methods, to samples with a sucrose:fructan ratio of 3:1 or 4:1 (or less; 8, 9). Methods based on determination of fructan from only fructose require two correction factors: C fructan = k W k G (C F,f ) (2) q +1 q where q = average fructose to glucose ratio of fructans. For the special case of GF n fructans: � G = � �� avg �� ��� -1 � (3) which makes the overall correction ( k W k G ), in terms of DP avg : k G = correction for glucose content =

(4)

The method of Cuany et al., recognizes that, when sodium borohydride treatment is used to eliminate free reducing sugars, Equation 4 applies to fructans of both GF n and F m type because, if fructan is calculated on the basis of fructose only, the terminal glycosyl residue is lost for all forms, just as it would be for only GF n species (10). The shortcoming of this method, as with all methods to date, is the need to either have independent knowledge of the fructan ingredient (which would allow selection of a specific correction factor) or application of a common factor regardless of fructan type. The method presented here is based on the chemistry used in the Cuany method (10), but with a few distinct differences, the most important being the addition of a procedure for selecting the appropriate correction factor to be applied for calculating the total fructan content and significant procedural simplification by eliminating the need for SPE cleanup.

Table 2016.06A. Total fructan single-laboratory validation data: precision Sample type No. of replicates Mean, g/100g, RTF a SD r

, % SD IP b

, % c

RSD r

RSD IP

Materials from SPIFAN sample kit Child formula powder, placebo Toddler formula powder, milk based Infant formula powder, milk based Infant formula powder with FOS/GOS e Adult nutritional RTF high fat Abbott Nutrition in-house materials Infant formula powder, soy based Child formula powder

12 d 12 d 12 d 12 d 12 d 12 d 40 f 40 f 40 f 60 g

0.270 0.233 0.283 0.277 0.036 0.500 0.153 0.434 0.230 0.370

0.0055 0.0080 0.0059 0.0072 0.0008 0.0184 0.0025 0.0060 0.0052 0.0040

2.03 3.42 2.09 2.61 2.14 3.67 1.64 1.38 2.24 1.09

0.01547 0.00806 0.00696 0.01238 0.00117 0.03395 0.00442 0.01167 0.00892 0.00951

5.73 3.46 2.46 4.47 3.29 6.79 2.89 2.69 3.88 2.57

Adult nutritional powder

Pediatric powder, milk based Control powder, milk based

a RTF = Ready-to-feed. b SD IP

= Standard deviation (intermediate precision).

c RSD IP = Relative standard deviation (intermediate precision). d Duplicates on each of 6 days, one laboratory. e FOS/GOS = Fructooligosaccharides/galactooligosaccharides. f Duplicates on each of 10 days, in each of two laboratories. g Duplicates on each of 10 days, in each of three laboratories.