376 - Impact of a nationwide blood culture bottle shortage on antibiotic use among hospitalized pediatric patients with fever

Publication Number: 4368.376

Shirley Zhao, Yale School of Medicine, Hamden, CT, United States; Frances Y. Cheng, Yale University School of Medicine, New Haven, CT, United States; David R. Peaper, Yale School of Medicine, New Haven, CT, United States; Thomas Murray, Yale School of Medicine, New Haven, CT, United States; Morgan Harper, Yale School of Medicine, Hamden, CT, United States; Veronika Shabanova, Yale School of Medicine, Hamden, CT, United States

Background: In June 2024, hospitals nationwide experienced a severe shortage of BD Bactec™ blood culture bottles. The shortage, which led to an 80% supply reduction at our institution, prompted health systems to rapidly implement stewardship strategies to conserve culture availability. The clinical impact of the shortage on pediatric patients is unknown.

Objective: To evaluate the impact of a nationwide BCx bottle shortage on antibiotic use among pediatric inpatients within a tertiary health system with fever.

Design/Methods: We conducted a retrospective cohort study comparing patients admitted during (06/28/2024-10/14/2024) and after (10/15/2024-01/31/2025) the BCx bottle shortage to those admitted during the same periods one year prior. Inclusion criteria were patients ages 0-21 years admitted after presenting to the emergency department with a chief complaint of fever or a documented temperature ≥ 100.4 °F. The primary outcome was antibiotic use (proportion of hospital days with systemic antibiotic administration). Secondary outcomes included the proportion of admissions with any BCx obtained, median BCx sets (i.e. set of 1 or 2 bottles collected from a peripheral or central site) per admission, positivity rates (i.e. proportion of BCx sets with a positive result), median length of stay (LOS), and 30-day readmission rates. Chi-square and unpaired Mann-Whitney tests were used for statistical analysis.

Results: Of 1717 admissions, 44.8% were female and median age was 3.7 years (IQR 1.2-9.7) (Table 1). During the shortage, antibiotic use was unchanged from the year prior; however, post-shortage, the proportion of days with antibiotic administration increased significantly (71.2% vs 64.0%, p< 0.0001). BCx use decreased significantly during and post-shortage, including proportion of admissions with any BCx obtained (shortage: 43.2% vs 56.7%, p< 0.0001; post-shortage: 41.6% vs 50.2%, p=0.005) and mean BCx sets per admission (shortage: 0.56 vs 0.92, p=0.005; post-shortage: 0.58 vs 0.78, p=0.009) (Table 2). Median LOS and 30-day readmission rates were unchanged across timeframes. Sub-analysis of admissions by service showed similar findings among pediatric hospitalist admissions. Febrile infants (age ≤ 60 days) had no significant difference in outcomes.

Conclusion(s): At our institution, the BCx bottle shortage was associated with increased antibiotic use post-shortage and decreased BCx use during and after the shortage. Pediatric hospitalist service patients were similarly impacted, while febrile infants were not. These findings illustrate how BCx supply disruptions can impact diagnostic and management decisions.

Table 1. Demographic characteristics of pediatric patients admitted during study timeframes
Demographic characteristics of pediatric patients admitted during the blood culture bottle shortage, post-shortage, and their historical controls. P-values calculated using Chi-square and unpaired Mann-Whitney tests.

Table 2. Outcomes of blood culture bottle shortage among pediatric patients hospitalized with fever
Impact of blood culture bottle shortage on antibiotic days, blood culture use, length of stay, and 30-day readmission rates among pediatric patients hospitalized with fever. P-values calculated using Chi-square and unpaired Mann-Whitney tests.