Baseline and seasonal trends of Bacillus cereus and Bacillus subtilis from clinical samples in Japan.

Background: Outbreaks of Bacillus cereus bloodstream infections (BSIs) are a concern in Japanese medical settings. Aim: This study determined baseline values for B. cereus detection in clinical samples that are useful as reference values for hospitals when assessing the need for intervention. Method: A retrospective analysis of B. cereus detection in the Japan Nosocomial Infections Surveillance data from 2008 to 2014 was performed; it included 950 individual hospitals across the country. Findings: Bacillus spp. were detected in 0.54% of the clinical specimens submitted for bacteriological testing. Specimens positive for Bacillus spp. were mainly blood (24.6%), stool (26.5%), and respiratory specimens (23.3%). Identification of Bacillus spp. at the species level (i.e., B. cereus or B. subtilis) was reported in 55.3%, 14.7%, and 15.4% of cases, of which 88.9%, 48.3%, and 33.1% were B. cereus in blood, stool, and respiratory specimens, respectively. Of the 4105 hospital-years, 75.7% had blood specimens with Bacillus spp., with a median of 0.85 blood specimens/100 beds annually (interquartile range, 0.17-2.10). The B. cereus detection showed significant summer seasonality, regardless of specimen type or geographic distribution. The B. subtilis detection did not show seasonality, and its detection remained constant throughout the year. The seasonality of Bacillus spp. reflects the high proportion of B. cereus. Conclusions: The increased detection rate of Bacillus spp. during summer should be interpreted as a risk factor for B. cereus BSIs. A post-summer decrease in Bacillus spp. should not be interpreted as an effect of interventions.