Although Helicobacter infections of laboratory mice are usually subclinical, they may interfere with in vivo experiments and thus may lead to misinterpretation of data. As such, it is important to provide a means to unequivocally identify infections with murine Helicobacter spp. In the present study, a nested polymerase chain reaction (PCR) was established and shown to be 10 to 100 times more sensitive than the single-step PCR commonly used for routine diagnosis of Helicobacter spp. Experimental infection of Helicobacter-free mice demonstrated that faeces, caecum, colon and rectum but not liver are equally suitable for the detection of H. bilis. However, use of faecal pellets is advantageous since detection of H. bilis is possible one week after infection and analysis of faeces instead of tissues avoids euthanasia of animals. Furthermore, it generates representative data for all animals housed in the same cage and analysis can be repeatedly performed. Use of samples from breeding pairs but not offspring provides representative information about the Helicobacter status of a mouse colony. Both C3H/HeJ and C57BL/6 mice appear to be susceptible to H. bilis and persistent infection was observed during the 20-week experimental period. Analysis of pooled faecal pellets by nested PCR seems to be the most sensitive approach for H. bilis monitoring of the given breeding colony.