Contamination of naturally gluten-free grains
My colleagues (Anne Lee, Schar, USA and Thomas Grace, Bia Diagnostics) and I recently published findings from a pilot study on grain contamination in the June issue of the Journal of the American Dietetic Association (Thompson T, Lee AR, Grace T. Gluten Contamination of Grains, Seeds, and Flours in the United States: A Pilot Study. J Am Diet Assoc. 2010;110:937-940). What follows is a summary of study findings.
Bottom line: Seven of 22 (32%) samples tested contained mean gluten levels above 20 ppm with amounts ranging from 25 to 2,925 ppm.
Why we conducted this study: In the FDA’s proposed rule for labeling of food as gluten free, single ingredient foods, such as corn, rice, and millet are considered inherently gluten free. These grains will be considered misbranded if they carry a gluten-free label that does not also state that all foods of that same type are gluten free (e.g. “all millet is gluten free” or “millet, a gluten-free food”).
Oats are the only grain that will not be considered misbranded if they carry a gluten-free label and do not also state that all foods of the same type are gluten free. In fact, because research suggests that commercially available oats may be contaminated with wheat, barley, and rye, labeled gluten-free oats will be considered misbranded if the label implies that all oats are gluten free.
Unfortunately, oats may not be the only naturally gluten-free grain contaminated with gluten.
What we tested: Twenty-two naturally gluten-free grains, seeds, and flours NOT labeled gluten-free were tested in duplicate using the Ridascreen Gliadin sandwich R5 ELISA with cocktail extraction. Seven of the 22 products contained a voluntary allergen advisory statement for wheat; 15 did not. Products included white rice and flour, brown rice, corn meal, polenta, buckwheat and buckwheat flour, amaranth seed and flour, flax seed, millet grain and flour, sorghum flour, and soy flour.
Findings: Thirteen of 22 (59%) products contained below the limit of quantification for gluten which is 5 ppm for the assay used. Of these 13 products, 3 contained a voluntary allergen advisory statement for wheat. Nine of 22 (41%) products contained more than the limit of quantification for gluten, with mean gluten levels ranging from 8.5 to 2,925. Of these nine products, four contained a voluntary allergen advisory for wheat.
Conclusions: Results of this study confirm that a certain percentage of inherently gluten-free grains, seeds, and flours are NOT gluten-free when they are purchased by consumers. Co-mingling of grain and seed can occur anywhere along the line from the field to the packaging plant.
Results also suggest that consumers can not rely on voluntary allergen advisory statements for wheat to make decisions about which products are more or less likely to be contaminated. Four of seven products containing greater than or equal to 20 ppm gluten did not contain an allergen statement for wheat while three of the products that contained below the limit of quantification for gluten did contain an allergen advisory statement.
While we can infer from this study that some degree of contamination exists in naturally gluten-free grains, seeds, and flours sampling was not large enough to make any assessment on the overall percentage of contaminated product.
Sampling also was not large enough to make any inferences on the specific grains, flours, and seeds more or less likely to be contaminated.
The following products contained above the limit of quantification for gluten:
White rice flour/8.5
The following products tested below the limit of quantification for gluten: basmati rice; long grain brown rice; enriched corn meal; instant polenta; 1 sample of rice flour; hulled buckwheat; buckwheat groats; amaranth flour; flax seed; and amaranth seed.
For more information about this study, please contact me at firstname.lastname@example.org
This study was funded in part by Schar USA, a manufacturer of prepared gluten-free foods
The study abstract is available on Pub Med http://www.ncbi.nlm.nih.gov/pubmed/20497786
Copyright © by Tricia Thompson, MS, RD
Also available at: