Biofilms are the most common mode of bacterial growth in water distribution systems, causing pipe overgrowth and possibility of pathogen survival. Traditionally pathogenic bacteria are detected by culture methods, however, only a small part of bacteria associated with water can be cultured. Thus there still is a need for information on fate of pathogens in water distribution systems. The survival of Escherichia coli was investigated using a biofilm reactor (Propella TM). A pure culture of E. coli bacteria was inoculated into non-sterile, E. coli free synthetic water and incubated for 96 hours. Each day biofilm and water samples were analysed for E. coli using a traditional plate count method, fluorescent in situ hybridization (FISH) combined with Direct Viable Count (DVC) and QRT-PCR. Results showed loss of culturability within 90 hours, however, using DVC-FISH method after 96 hours of incubation viable cell numbers had decreased for only 1 log. Analysis between the normal distribution of cell length and residence time showed decrease in average DVC positive cell length with incubation time. Analysis with FISH and QRT-PCR gave similar results on E. coli increase in reactor biofilm (by more than 1 log) and decrease in water (by 1 log) during the experiments. By combining FISH positive cell numbers in biofilm and water 0.03 h-1 cell growth of E. coli was obtained with main source of growth being the biofilm. In conclusion, it was shown that E. coli even in oligotrophic conditions can accumulate in biofilm and show potential to divide even without being detectable with traditional culturing methods.