An interlaboratory study was performed in eight laboratories to validate an ELISA method developed for quantitative determination of casein in foods. The ELISA kit used is based on rabbit polyclonal antibodies. The kit is quite specific; no false-positive results or cross-reactivities were obtained for a broad range of food matrixes with zero content of milk proteins. All participants in the study received the casein kit, which included a standard operating procedure, a list of the samples, the samples, and a protocol for recording test results. The study included nine food samples: wheat flour, buckwheat flour, instant potato purée with milk, instant coffee with sugar and cream, a mixture for fancy bread, salami, liver paté, chocolate muesli with nuts, and a mixture for gluten-free bread. Three food samples with zero content of milk proteins showed a casein content lower than the lowest casein standard (1.0 mg CAS/kg) in most laboratories and measurements (64%). In 98% of the cases, the casein content was lower than the estimated LOQ. Two food samples with no dairy ingredient declared on the ingredient list contained casein levels higher than the second casein standard (3.0 mg CAS/kg) and the third standard (10.0 mg CAS/kg), respectively. Four food samples containing milk as an ingredient tested positive, and three showed casein contents higher than the highest standard (30.0 mg CAS/kg). The statistical tests (Cochran, Dixon) and analysis of variance were used for evaluation of the interlaboratory study results. Repeatability and reproducibility limits as well as LOQ (1.8 mg CAS/kg) and LOD (0.5 mg CAS/kg) for the kit were calculated from the results of the interlaboratory study.

SEDIUM RD Ltd Zeleznicního pluku 1361 530 02 Pardubice Czech Republic