Abstract
BACKGROUND: Bacillus cereus can occasionally cause invasive infections, irrespective of the host's immune status. However, the presence of B. cereus in a single blood culture is typically considered a contaminant, potentially leading to an underestimation of the true incidence of B. cereus bacteremia. This study aimed to evaluate the cut-off value for time to positivity (TTP) to distinguish true bacteremia from contamination. METHODS: This single-center retrospective study included patients with positive blood cultures for B. cereus between November 2011 and February 2025. A receiver operating characteristic (ROC) curve was generated to evaluate the diagnostic performance of TTP, which was achieved by plotting the sensitivity (the true-positive rate) on the y-axis against 1 - specificity (the false-positive rate) on the x-axis for a series of potential TTP cut-off points. RESULTS: During the study period, 49 patients were enrolled, of whom 18 were classified into the true bacteremia group and 31 into the contamination group. All patients had at least two sets of blood cultures collected. The median TTP was significantly shorter among patients in the true bacteremia group (7.8 h; interquartile range [IQR], 6.5-8.9) than in the contamination group (11.2 h; IQR, 9.9-13) (p < 0.001). ROC curve analysis indicated that a cutoff value of 9.0 h achieved the highest sensitivity (77.8%) and specificity (90.3%) for predicting true bacteremia, while a more lenient threshold of TTP > 13.0 h yielded 100% sensitivity for identifying contamination. CONCLUSIONS: A cut-off value of 9.0 h can differentiate true B. cereus bacteremia from contamination, whereas a TTP > 13 h is indicative of contamination. Given its suboptimal diagnostic accuracy, TTP can help differentiate true B. cereus bacteremia from contamination but should be interpreted with clinical and laboratory data for the diagnosis of sepsis.