To systematically evaluate the applicability of adenosine triphosphate fluorescence method in the commercial aseptic rapid detection of dairy products.

The detection ability of 35 kinds of representative microorganisms was tested by the inclusion experiment using the adenosine triphosphate fluorescence method. The anti-interference performance of adenosine triphosphate fluorescence method in 70 dairy product matrices of 7 types was evaluated through exclusive experiments. Through artificial contamination experiments, using the plate streptography method and GB 4789.26—2023 National food safety standard-Microbiological testing of food-Commercial sterility testing as reference methods, the detection sensitivity of 7 kinds of representative contamination bacteria in 7 types of dairy products was systematically compared. The 2 laboratories were arranged to perform inter-laboratory comparison tests for evaluating the reproducibility of the established method.

The inclusive experiment and exclusive experiment results of the adenosine triphosphate fluorescence method were both 100%. In the sensitivity test, the limit of detection at 50% positive rate (LOD₅₀) of the developed method for Lactobacillus plantarum was 0.32 CFU/ sample, slightly superior to the streak plate method (0.37 CFU per sample) and the national standard GB 4789.26—2023 (0.37 CFU per sample). The LOD₅₀ values for the other 6 kinds of target bacterial strains were basically consistent with those of the reference methods, with the relative limit of detection(RLOD) approaching 1. Interlaboratory comparison results verified favorable reproducibility of this method, as the RLOD for most strains were close to 1. This method could achieve early detection of Aspergillus niger, Enterococcus faecalis and other microorganisms following 3 days of incubation. In comparison with GB 4789.26—2023, which required a detection period of 7–10 days, the assay duration was reduced by 4–7 days.

The adenosine triphosphate fluorescence method features rapid detection speed, simple operation, as well as good inclusivity and specificity. Its sensitivity is comparable to that of traditional methods, making it suitable for the rapid screening and process monitoring of commercial sterility in dairy products. This study provides experimental evidence for shortening the detection cycle of commercial sterility in dairy products and promoting the application of rapid detection technologies. Meanwhile, it constructs a reusable methodological validation framework, providing a reference for the validation of other rapid detection technologies in food matrices, and offers experimental evidence for the “primary screening-rechecking” process in practice.