Pseudo-outbreak of toxigenic Bacillus cereus isolated from stools of three patients with diarrhoea after oral administration of a probiotic medication.

From December 2000 to January 2001 toxigenic Bacillus cereus was isolated from stools of three patients with diarrhoea at two tertiary hospitals in southwest Germany. Two cases with nosocomial diarrhoea were apparently epidemiologically related (same time and ward), a third case was unrelated with respect to time and location. In order to investigate the epidemiology of these three cases, clinical isolates and isolates from an unexpected, possible common source (probiotic medication) were compared by toxin assay, biotyping and randomly amplified polymorphic DNA (RAPD) analysis. The three clinical isolates, as well as the two isolates from different lots of the probiotic medication (Bactisubtil containing 'Bacillus IP 5832'; Cassella-med, Cologne, Germany), were indistinguishable by toxin assay, biotyping and RAPD, when compared with other distinguishable clinical B. cereus strains. As the diarrhoeal disease had begun before the probiotic medication had been administered to overcome it, the isolated B. cereus probably was at least initially, not the cause of the observed diarrhoeal disease. Isolation of toxigenic B. cereus from stools appeared to be a diagnostically misleading epiphenomenon after oral medication with the probiotic. We conclude, that probiotic medication with Bactisubtil (Bacillus IP 5832) may result in diagnostically misleading results when culturing stool specimens from patients with diarrhoea. The clonal identity of isolates may be misinterpreted as an outbreak. Stool specimens should be taken before start of probiotic treatment and clinicians should state probiotic medication when ordering stool examinations to allow correct interpretation of results. Nevertheless, it is noteworthy that a probiotic medication contains potentially toxigenic material.