Management of a positive blood culture in the pediatric emergency department: a multicenter case-based survey.

Abstract

Objective: Positive blood cultures in pediatric emergency departments (ED) represent a diagnostic dilemma for clinicians as 50% are contaminants. Our goal was to characterize management of positive blood cultures by physicians working in pediatric EDs across Canada.

Methods: A self-administered electronic survey was sent to all members of the Pediatric Emergency Research Canada (PERC) network in 2024 using the principles of the Dillman's tailored design method. The survey was constructed using standardized methodology. Potential risk factors of bacteremia were identified based on a literature review and consultation of multiple experts. The survey consisted of three clinical cases, modeled off real patients, and was pilot tested by a group of physicians. Each case highlighted a common clinical scenario in which physicians had to interpret the significance of a positive blood culture. There were follow-up questions to further assess clinical decision-making and provide demographic information.

Results: In the final analysis, we included 153 (69%) of the 221 PERC members invited from 15 hospitals. The management of the case scenarios was heterogeneous, for example, 49% of participants would have discharged a 9-week-old with a positive culture and otherwise normal blood results while 35% suggested hospitalization with intravenous antibiotics. Most participants suggested that young age, immunocompromised status, and shorter time to blood culture positivity as risk factors for true bacteremia, though there was no consensus on time-to-positivity or influenza status. 98% of the participants reported that they would be willing to adopt a clinical decision rule if it had a high sensitivity.

Conclusion: There is considerable practice variation among ED physicians for the management of positive blood cultures in children across Canada. Incongruencies in the perceived impact of clinical factors on the likelihood of a true bacteremia, such as time-to-positivity, highlight the need for a standardized decision-making tool.