Introduction
Obtaining blood cultures is an essential practice for the diagnosis of bloodstream infection; drawing at least 2 sets of blood cultures is conventionally recommended for diagnosing bloodstream infection and identifying causative pathogens.
The Becton Dickinson announced a potential supply shortage of BACTEC® blood culture bottles in June 2024. Reference Beddard1 In response, the American Society for Microbiology published a guideline to manage the situation, which provided guidance on how to maintain the diagnostic yield of blood cultures through measures such as optimizing blood sampling volume and timing, minimizing blood culture contamination and prioritizing blood culture practice in the context of diagnostic stewardship. Reference Suleyman, Moore and Palavecino2 The study hospital instituted a policy of collecting a single set of blood cultures as a temporary measure during the shortage period. The present study describes how the standard practice of collecting multiple blood cultures was restored after the resumption of the blood culture bottle supply.
Methods
The study was conducted at Fujita Health University Hospital, an academic tertiary-care hospital in Aichi, Japan. The proportion of blood cultures drawn in multiple sets was tracked as a quality measure during the following prespecified study periods: the pre-shortage period (from April to June 2024), the shortage period (from July to September 2024), and the recovery period (from October to December 2024). The single blood culture practice was defined as the blood culture obtained from a patient on a given day without multiple blood cultures collected on the same day. Contamination was defined as normal skin flora (eg, coagulase-negative staphylococci) that was only recovered from 1 set of multiple blood cultures. The proportion (%) of multiple blood cultures was determined as (total number of blood cultures drawn–total number of single blood cultures drawn) divided by the total number of blood cultures drawn ×100. The hospital-wide blood culture testing density per 1,000 patient days was also tracked.
The study hospital implemented a single-set blood culture policy across the hospital except for the hematology/oncology department during the shortage period. The policy change was announced via a pop-up notification in the electronic health records, at the hospital practice meeting, and through the hospital newsletter and hospital emails. Upon resolution of the shortage, effective from October 1, termination of the policy change and resumption of the 2-set blood culture practice were announced through the same venues. Blood culture practices at pediatric departments were excluded from the data tracking and analysis.
Results
The proportion of multiple blood cultures during the pre-shortage period ranged from 90% to 95%, while its proportion during the shortage period decreased to 7.1% in the second week. In the recovery period, the proportion of multiple blood cultures increased to 77% in the first week; however, its proportion remained below 90% during the following 3 weeks, prior to returning to greater than 90% in December (Figure 1). The weekly average testing density of blood culture practice per 1,000 patient days during the pre-shortage period, the shortage period, and the recovery period were 48.1, 28.1, and 49.5, respectively (Figure 1).
The proportion of multiple blood cultures and blood culture collection per 1,000 patient days (weekly).
The weekly average numbers of positive blood cultures representing true bloodstream infection during the pre-shortage period, the shortage period, and the recovery period were 46 (range, 28–71), 35 (range, 19–52), and 51 (range, 28–79), respectively. During the shortage period, 76 patients with a single set of positive blood cultures due to normal skin flora were subsequently identified, and 41 (54%) received antimicrobial therapy for true bloodstream infection.
Discussion
Although the rapid implementation of blood culture practice guidance during the blood culture supply shortage period was successful in minimizing a perturbation in clinical frontlines, the present study revealed that it took approximately 1 month for the multiple blood culture practice to return, but the proportion of multiple blood culture practice in the recovery period remained numerically lower than that in the pre-shortage period despite the hospital-wide announcement of blood culture shortage resolution.
The slow transition back to the 2-set blood culture practice may have negatively affected the diagnosis of bloodstream infection. Previous studies demonstrated that causative pathogens for bloodstream infection were detected in only 73.1%–85.7% of cases under the single-set blood culture practice. Reference Elantamilan, Lyngdoh and Khyriem3,Reference Lee, Mirrett, Reller and Weinstein4 Consistent with previous studies, the difference in the average number of positive blood cultures between the pre-shortage and shortage periods was 11, suggesting that up to 24% of positive blood cultures may have been overlooked during the shortage period. Moreover, in the single-set blood culture practice, the distinction of true bloodstream infection from contamination for skin commensal flora would be cumbersome, Reference Doern, Carroll and Diekema5 which may have led to antimicrobial overuse. As seen in the results, over half of patients with single-set positive blood cultures due to normal skin flora during the shortage period received antimicrobial therapy.
The present study has several limitations. Data was collected from a single hospital, and the changes in practice patterns may not apply to other institutions. We did not assess the volume of blood inoculated or the timing of blood culture sampling as a quality measure, which may affect the proportion of blood culture positivity. Since the present study did not collect patient-level data, whether the shortage contributed negatively to patient clinical outcomes is yet to be ascertained.
Shortage of blood culture bottle supply appears to have detrimental effects on clinical practice, and restoring optimal blood culture practice may be delayed for some time following the restoration of the supply. Consistent monitoring of objective blood culture practice data would be crucial as a quality indicator, and it would also offer a great opportunity to reconsider the optimal blood culture practice.
Data availability statement
We will share the data underlying this article with the corresponding author upon reasonable request.
Acknowledgments
None.
Author contribution
H.H. designed the study. C.S. and Y.H. obtained the data. H.H. and S.H. analyzed and interpreted the data. S.H. drafted the article. Y.U., Y.D., and H.H. revised the manuscript. All the authors critically reviewed the text and approved the final version of the manuscript.
Financial support
This study received no financial support. Y.D. was supported in part by the National Institute of Health (R01AI104895).
Competing interests
Y.D. received consulting fees from GSK, Moderna, Gilead, Fujifilm, Shionogi, Meiji Seika, Pfizer, and AbbVie; honoraria from Gilead, Shionogi, and BD; and research funding from Entasis, which were unrelated to the work. H.H. received honoraria from Shionogi, Moderna, bioMérieux, Pfizer, Takeda, and Pfizer and consulting fees from Solventum and Moderna, which were unrelated to the work. Y.U. received grants from Kao and honoraria from Shionogi, MSD, Pfizer, and BD.
All the other authors report no conflicts of interest relevant to this article.
Ethical standard
The institutional review board at Fujita Health University approved the study (HM24-397).
