Usefulness of bifidobacteria for the detection of faecal contamination in milk and cheese

Research of faecal contamination is most often based on the research of Escherichia coli, a rather unspecific marker showing both survival and multiplication outside the intestinal tract. When researching a more specific and sensitive marker, we examined human and cow's faeces, raw milk and dairy products for the presence of E. coli, but also bifidobacteria. In 146 human faecal samples, bifidobacteria were present at a mean level of 7.7 log cfu.g(-1), and the identified species were B. bifidum, B. longum, B. adolescentis and B. infantis. E. coli was present at a level of 6.9 log cfu.g(-1). In 30 samples of cow's faeces, the highest count of bifidobacteria was 8 log cfu.g(-1), while E. coli was present at a level of 6 log cfu.g(-1). Among 15 samples of cheese prepared with raw milk, bifidobacteria were found in 14 samples at levels comprising between 2 and 5 log cfu.g(-1). E. coli was never found (detection level 2 log cfu.g(-1)). Analysis of 207 samples of raw cow's milk sampled on the teat allowed isolation of bifidobacteria in 90% and E. coli in 68% of samples. Among 66 samples of raw milk sampled in refrigerated tanks, bifidobacteria were isolated in 95% and E. coli in 86%. In all cow's samples (faeces, milk and cheese), one distinct species was identified: Bifdobacterium pseudolongum var globosum. Levels of bifidobacteria are higher than values of E. coli in both human and animal faeces and raw milk, underlining the choice of bifidobacteria as a valid marker for faecal contamination of dairy products. Identification of bifidobacterial species even allows the determination of the origin (animal or human) of contamination.