作者: I Jongenburger , MW Reij , EPJ Boer , LGM Gorris , MH Zwietering
DOI: 10.1016/J.IJFOODMICRO.2010.07.025
关键词:
摘要: This study aims to assess several factors that influence the accuracy of plate count technique estimate low numbers micro-organisms in liquid and solid food. Concentrations around 10CFU/mL or 100CFU/g original sample, which can still be enumerated with technique, are considered as numbers. The impact counts, technical errors, heterogeneity contamination singular versus duplicate plating were studied. Batches powdered milk artificially contaminated various amounts Cronobacter sakazakii strain ATCC 29544 create batches accurately known levels contamination. After thoroughly mixing, these extensively sampled plated duplicate. coefficient variation (CV) was calculated for samples from both product a measure dispersion within samples. errors counts determined theoretically, experimentally, well Monte Carlo simulations. CV-values found similar their theoretical established by assuming Poisson distribution counts. However, approximately five times higher than CV-values. In particular, spp. showed much more expected likely due heterogeneity. less prominent Considering on accuracy, it would advisable keep lower limit 25 colonies/plate rather currently advocated 10 colonies/plate. For heterogeneous contamination, is accurate use plates individual same 5