The hygiene of chicken processing surfaces and retention of the wheat protein gliadin and of protein in general on those surfaces were compared in 15 trials after 3 increasingly rigorous cleaning steps. Eleven different chicken products with wheat derivatives as a batter were prepared on 3 processing lines in 15 production runs selected at random over 6 mo (5 runs were thus replicates). Using surface swabs, surface hygiene was monitored by adenosine triphosphate (ATP) bioluminescence, gliadin by immunoassay, and protein by the Coomassie dye method. Gliadin was monitored in 14 trials, protein in 5, and all trials were monitored by ATP bioluminescence. In a typical trial, gliadin values normalized to uncleaned values fell from 100000 arbitrary units, to 6000 after rinsing, to 30 (foam, rinse), to not detected (sanitize, rinse). Parallel ATP bioluminescence values also decreased, but crucially, the relative gliadin value was less than the relative ATP value after foam and rinse in all 14 trials, a result unchanged after sanitize and rinse. In trials comparing ATP and protein, the relative ATP values exceeded the relative protein values in 4 of 5 trials after foaming and after sanitizing. Thus, for these 11 products, ATP bioluminescence was a surrogate indicator of residual gliadin and probably of residual protein. Absolute gliadin concentration on an uncleaned processing line was also the basis of modeling the risk of cross-contamination of gliadin in follow-up product, where the line was hypothetically left uncleaned between production runs. The results show that all follow-up product could be declared "gluten-free" under proposed legislation, and suggest that some industrial cross-contamination risks are currently overestimated.