Abstract
Blood culture is the main tool used to identify causative pathogens. Adequate volume and number of culture sets are considered key to blood culture positivity rate. It is not known whether these factors remain critical to the positivity rate after the introduction of automated continuous blood culture system monitoring. We measured blood volume per bottle and described the distribution of blood volume and number of culture sets. Multivariate logistic regression was performed to determine the independent association of blood volume, number of culture sets, diagnosis of sepsis in a patient, and other covariates with blood culture results. Only 6.9% of the blood culture bottle volumes complied with the guidance (8-10 mL), with the highest culture positivity rate (18%). Of the culture events, only one set of blood was cultured in 60.9% of events. In the multivariate analysis, blood culture volume per event (odds ratio [OR], 1.09 [95% confidence interval [CI], 1.06-1.11]), patients with a diagnosis of sepsis (OR, 2.86 [95% CI, 2.06-3.98]), and samples from the emergency department (OR, 2.29 [95% CI, 1.72-3.04]), but not the number of culture sets (OR, 0.74 [95% CI, 0.50-1.12]), were observed to be statistically significant with respect to blood culture positivity rate. Our results revealed that the total blood culture volume and the diagnosis of sepsis were critical factors affecting blood culture positivity rate. However, the proportion of blood culture bottles with the optimal blood volume was very low, and optimizing blood volume would be key to increasing blood culture positivity rate.