Every Crisis Is an Opportunity: Advancing Blood Culture Stewardship During a Blood Culture Bottle Shortage

Abstract

The current manufacturing disruption of BACTEC blood culture bottles has drawn attention to diagnostic stewardship around blood culture utilization. In this perspective, we offer strategies for implementing blood culture stewardship using a graded approach based on a hospital's blood culture bottle supply. These strategies should inform plans to mitigate the impact of the shortage on patient care and reinforce fundamental principles of blood culture stewardship.

The current blood culture bottle shortage emphasizes the importance of diagnostic stewardship for blood cultures. We offer strategies for implementing blood culture stewardship using a graded approach based on supply to mitigate impact and reinforce fundamental blood culture stewardship principles.

On 11 June 2024, BD Diagnostics (Becton, Dickinson and Company) alerted users of their BACTEC blood culture (BCx) system regarding expected shortages in BCx bottles due to supply chain disruptions [1]. This shortage affects BD BACTEC aerobic, anaerobic, pediatric, and fungal/mycobacterial BCx bottles [1]. Subsequently, on 10 July, the United States (US) Food and Drug Administration (FDA) added BCx media bottles to the medical device shortages list and published a letter to healthcare providers urging BCx “conservation strategies” [2]. On 23 July, the Centers for Disease Control and Prevention (CDC) released a Health Advisory urging healthcare providers, laboratory professionals, healthcare facility administrators, and health departments affected by this shortage to begin to assess their situations and implement mitigation plans immediately [3]. This shortage affects a large number of hospitals: Approximately half of US laboratories use BD BACTEC systems, and 59% of >1000 participants of a recent call hosted by the CDC, Infectious Diseases Society of America, Society for Healthcare Epidemiology of America, Pediatric Infectious Diseases Society, and American Society for Microbiology (ASM) on 23 July reported <1 month’s supply of BCx bottles [4, 5].

In this perspective, we provide a practical framework for implementation of strategies to optimize BCx use using a graded approach based on supply levels (conventional, crisis, and contingency capacity) with specific implementation examples. This document is complementary to a recently published paper by ASM on managing BCx shortages [6]. We discuss BCx best practices, address challenges with implementing BCx stewardship programs, and offer potential solutions. Our hope is this shortage serves as an opportunity to implement BCx stewardship as a conventional strategy that will be maintained after the BCx bottle shortage is over.

A GRADED APPROACH TO ADDRESS THE BLOOD CULTURE BOTTLE SHORTAGE

Healthcare institutions are experiencing differing impacts on BCx bottle supplies [3]. Using the framework developed by the National Institute for Occupational Safety and Health for preserving personal protective equipment during the COVID-19 pandemic, we outline BCx stewardship strategies based on BCx bottle capacity levels (conventional, contingency, and crisis) [7]. Conventional capacity strategies are considered best practices; therefore, healthcare institutions should aim to implement these at baseline and continue even after the shortage has abated. Contingency and crisis capacity interventions deviate from best practices and are temporarily implemented when supplies are inadequate. These categories are outlined with examples of strategies in Table 1.

Table 1.

Stepwise Strategies to Mitigate Blood Culture Bottle Shortage Based on Capacity

Strategy	Advantages	Disadvantages	Estimated Impact on BCx Utilization
Conventional capacity (strategies that should be implemented even when BCx bottle supply is adequate, as they are considered best practices)
BCx stewardship using indication-based algorithms	Evidence-based strategy to reduce unnecessary BCx Standardizes practices Resources already available for adaptation/implementation [8] Applicable to multiple settings (eg, ED, inpatient) and patient populations including adults and pediatrics	Algorithms are complex with multiple steps and clinical nuance Requires clinician education and ideally EMR integration (paper-based algorithms usually less preferred)	High impact (20%–50% reduction) Most effective when implemented with other strategies such as feedback
BCx stewardship around collection including optimizing volume, number of sets, and reducing BCC	Reducing BCC reduces repeat BCx, antibiotic use, and healthcare resource utilization Adequate volume increases BCx yield 2 sets improve BCx yield	Minimal impact if already low BCC Requires education/training of phlebotomy and nursing staff	Variable impact: dependent on current practices May be substantial if BCC rates >3% and/or BCx volumes are low
Contingency capacity (strategies that are not typically considered standard of care, but are temporally implemented as benefits outweigh harm when supply is short)
Decrease BCx utilization where primary infection site is available for culture and has greater yield than BCx Examples [8]: Urine cultures for pyelonephritis Bile cultures for cholangitis Sputum cultures for severe CAP Cultures of purulence for purulent cellulitis	Reduces BCx utilization while improving the opportunity to identify a pathogen	Primary source might not be readily available for culture (eg, bile) Antibiotic use may reduce yield Requires education Must risk-stratify implementation as detection of BSI may alter management (eg, Staphylococcus aureus bacteremia due to skin abscess in patient with prosthetic heart valve)	Moderate impact: may reduce BCx utilization for several common conditions such as pneumonia and pyelonephritis
Single BCx set for documenting clearance of bacteremia (see manuscript for discussion of examples)	Rapid reduction in use Many patients needing clearance of bacteremia are managed by infectious diseases consultants (smaller group to educate and better understanding of implication of missing bacteremia)	Reduced sensitivity of detection May be challenging to de-implement after shortage resolves	Low-moderate impact: depends on patient mix as primary impact likely to be with S aureus bacteremia and BCC
Cease using BCx bottles for body fluid cultures (eg, peritoneal fluid, pleural fluid)	Decreases BCx bottle utilization	Decreased yield from body fluid cultures Lab procedures need adaptation	Low impact: body fluid cultures utilize a minority of BCx bottles
Use fungal BCx when candidemia is suspected	Preserves bacterial BCx bottles	Very specific clinical circumstances limits impact	Low impact
Crisis capacity (strategies that significantly deviate from best practices/standard of care and are temporarily implemented due to unforeseen circumstances).
Single BCx sets for initial bloodstream infection detectiona	Rapid reduction in BCx use Changing clinical decision support can be enacted rapidly as indications do not change	Reduced sensitivity especially for Pseudomonas, Candida, and Enterococcus [9]; ∼10% of BSIs were missed due to a single aerobic bottle or no anaerobic bottle [10] May alter practice even after shortage resolves	High impact: reduces utilization significantly, but degrades pathogen detection ability
Use of pediatric BCx bottles	Additional supply may be available due to less frequent use Bottles can recover bacterial pathogens	Only hold 5 mL of blood per bottle, so requires twice as many bottles for equivalent volume	Low impact: stock likely not high enough to provide sustained use, particularly as also on shortage
Use of fungal BCx bottles	Additional supply may be available due to less frequent use Bottles can recover bacterial pathogens May be used for documenting clearance of candidemia	Fungal bottles not validated for recovery of bacterial pathogen	Low impact: stock likely not high enough to provide sustained use, particularly as also on shortage
Use of only anaerobic bottles	May allow for mitigation when imbalanced shortage of aerobic BCx bottles Yield remains high for many organisms (eg, S aureus) Can limit use to pathogens requiring repeat BCx where pathogen is known	Reduced sensitivity for detection of certain bacteria (eg, Pseudomonas aeruginosa and Candida spp) May require more complex clinical decision support or laboratory process change	Low-moderate impact: limited situations for which this is applicable without reducing yield
Use of expired BCx bottles	Provides additional source of BCx Must meet quality control checks	Very limited data [11, 12] Requires FDA and manufacturer guidance, as this practice may be cited by accreditation agencies as noncompliant	Variable impact: depends on available stock of expired BCx bottles
Change to send-out of BCx to neighboring institution or reference lab that utilizes non-BD BACTEC BCx system	Provides additional source of BCx Can be used for all or a portion of BCx obtained	Logistical issues with transportation and reporting (eg, EMR and lab information system) Potential delays in reporting Depends on contractual relationships between labs	Moderate-high impact: partial to complete replacement of BCx system, but may result in delays to patient care
Implement a different BCx bottle system	BCx yield will be similar	Validation of new system may be slow Expensive May not be practical during temporary shortage	High impact: no need to change utilization practices

Abbreviations: BCC, blood culture contamination; BCx, blood culture, BSI, bloodstream infection; CAP, community-acquired pneumonia; ED, emergency department; EMR, electronic medical record; FDA, Food and Drug Administration.

^aWhile 1 set meets the Centers for Medicare and Medicaid Services SEP-1 blood culture criterion, it is not recommended in routine practice due to the lower sensitivity of single sets compared to 2 sets in adult patients.

There are no defined thresholds for when to implement contingency or crisis strategies for optimizing BCx supplies. We suggest initiation of contingency plans if optimization of conventional practices is insufficient to address BCx bottle shortage (<1 month’s supply without expected replenishment) and crisis plans if supply is <10 days with uncertain replenishment. Close monitoring of BCx utilization rates and communication with suppliers are needed to inform implementation/de-implementation of strategies. A summary approach to respond to a BCx bottle shortage and tips for implementing BCx stewardship interventions are described in Figure 1.

Figure 1.

Blood culture shortage response coordination and tips for implementing blood culture stewardship strategies. Abbreviations: BCx, blood culture; ED, emergency department; EMR, electronic medical record.

CONVENTIONAL CAPACITY

Diagnostic stewardship advocates for the right test, for the right patient, to prompt the right action and has been practiced by both laboratorians and clinicians to optimize patient outcomes and the use of healthcare resources [13]. Below, we discuss the rationale for implementing programs to improve BCx utilization based on indications as a conventional capacity strategy (ie, when BCx bottle supplies are adequate).

Stewardship of Bacterial Blood Culture Indications

Most (90%) BCx ordered in hospitalized patients do not grow an organism [14]. Retrospective adjudication estimates that 30%–60% of BCx in intensive care units (ICUs) and 50% of BCx on medicine wards are inappropriate [15, 16]. Several BCx stewardship programs have used evidence-based clinical decision support tools to guide medical reasoning on appropriate BCx indications in adult and pediatric populations, leading to 20%–50% reductions in BCx utilization without concerning safety signals while reducing antibiotic use and central line–associated bloodstream infections (BSIs) and improving bacteremia detection [6, 16–20]. The impact of these programs may vary depending on how recommendations are implemented (eg, how education is disseminated, whether providers receive feedback on inappropriate cases), and what the recommendations are (eg, types of patients excluded, criteria to repeat BCx) and on local practices (baseline differences in inappropriateness). Since the shortage was announced, healthcare systems have implemented strategies to reduce low-yield BCx based on a previously published BCx algorithm [8] with a reported 30%–40% reduction in overall hospital BCx utilization [11]. Low-yield BCx that should be minimized are summarized in Table 2 (additional information can be found in the articles by Suleyman et al and Fabre et al) [6, 8].

Table 2.

Clinical Situations Where Blood Cultures Are Low Yield and Generally Not Recommended in the Absence of Severe Sepsis/Septic Shock [8, 14]

Initial blood cultures

Isolated fever or leukocytosis without other signs/symptoms of infection

Postoperative fever within 48 h after surgery

Nonsevere (non-ICU) presentations of common infectious syndromes:  • Cellulitis  • Clostridioides difficile colitis/gastroenteritis  • Community-acquired pneumonia/aspiration pneumonia  • Diabetes-related foot infection  • Lower urinary tract infection (eg, cystitis, prostatitis)  • Nonvertebral osteomyelitis  • Uncomplicated intra-abdominal infections (cholecystitis, appendicitis, diverticulitis; exception: cholangitis)

Surveillance BCx when no clinical signs or symptoms of infection are present:  • Prior to starting immunosuppression  • Prior to procedure/surgery  • Prior to line placement or starting total parenteral nutrition  • Daily cultures while on extracorporeal membrane oxygenation  • Outpatient hematopoietic stem cell transplant recipients on steroids [21]

Fever or leukocytosis with known noninfectious or viral syndromes:  • Aspiration pneumonitis  • COPD exacerbation/interstitial lung disease flare  • Gastrointestinal bleed  • Myocardial infarction  • Pancreatitis  • Pulmonary embolism  • Substance-related withdrawal syndrome  • Trauma

Follow-up or repeat blood cultures

BCx <48–72 hours after previous negative BCx without change in clinical status

Repeat BCx for persistent fever or leukocytosis without significant change in clinical status

BCx for clearance of BSI in patients clinically improved and with source control and no concern for endovascular infection (excludes Staphylococcus aureus, Staphylococcus lugdunensis, Candida spp)

BCx follow-up of single positive BCx bottle positive for contaminant(s) in patients without prosthesis and immunocompetent

Repeated BCx in persistent neutropenic fever without clinical change [22, 23]

Other situations

Clinical instability not due to infection (eg, trauma, cardiac event, hemorrhage, neurologic event)

Majority of outpatient BCx [24]

Abbreviations: BCx, blood culture; BSI, bloodstream infection; COPD, chronic obstructive pulmonary disease; ICU, intensive care unit.

While most BCx stewardship studies have focused on nonneutropenic patients, there have been emerging data suggesting utility in the hematology-oncology population [21, 22, 25–27] and the emergency department (ED) [18, 28]; hospitals may find opportunities to improve BCx utilization in these settings [6].

Stewardship of Blood Culture Collection

Blood culture contamination (BCC) results in both antibiotic and BCx overuse [29] and remains a conventional BCx stewardship strategy. As a quality indicator, healthcare systems should seek to achieve a BCC rate ≤1% [30, 31]. Volume of blood collected is the key determinant of BCx sensitivity, so ensuring that 8–10 mL is collected for each adult aerobic and anaerobic BCx bottle should be part of every BCx stewardship program [30]. Detailed guidance on optimal BCx collection is provided by the manufacturer and has been published in scientific literature [6, 31, 32].

Stewardship of Fungal Blood Cultures

Fungal BCx are ordered infrequently, and the shortage of these bottles is not as dire. However, opportunities for stewardship of fungal BCx exist as outlined in several recent publications [33–35].

Challenges to Implementation of a Blood Culture Stewardship Program

Barriers to implementing a BCx stewardship program may include clinical and system factors. Clinical barriers include clinicians’ overestimation of the presence of infection and/or BSI as well as fear of missing BSIs even when BSI incidence is low [36, 37]. Certain care settings can also lend themselves to overuse (eg, BCx may be ordered rapidly in the ED or ICU before a comprehensive clinical evaluation) [38]. System barriers may include limited resources to implement electronic medical record–based interventions, lack of unit leadership support to improve BCx utilization, and frequent turnover of clinical staff. Additionally, there have been concerns raised about the Centers for Medicare and Medicaid Services’ Severe Sepsis and Septic Shock Early Management Bundle (SEP-1), which encourages clinicians to obtain BCx in low-yield situations such as noninfectious processes (eg, pulmonary embolism, acute pancreatitis, chronic pulmonary obstructive disease exacerbation) or infections with low probability of bacteremia [39]. Strategies to mitigate these barriers include securing unit leadership support and identifying unit champions to promote BCx best practices (serve as point of reference as new clinical staff join the team), incorporating education on BCx best practices and the safety of BCx stewardship into both initial and recurring education, reviewing clinical cases with providers to increase their confidence and comfort with BCx decision making, adapting existing BCx algorithms to local patient populations with input of local relevant stakeholders, and securing healthcare system leadership support as needed to obtain timely resources for program implementation (eg, information technology resources, improve unit engagement of program, disseminate education).

CONTINGENCY CAPACITY

Contingency capacity refers to strategies that are not typically considered standard of care but are temporarily implemented, as benefits outweigh harm when supply is short. Examples are outlined in Table 1 and discussed below. Tips for implementing BCx bottle conservation strategies are summarized in Figure 1.

Minimizing Blood Cultures for Infections Where the Primary Source of Infection Is Available for Culture

In syndromes where a culture from the primary source can be obtained quickly, BCx could be restricted during contingency. For example, urine cultures for pyelonephritis and respiratory cultures for severe community-associated pneumonia are much more likely to yield a pathogen than BCx [40, 41]. Additional examples are shown in Table 1.

Single Blood Culture Sets to Document Clearance of Bloodstream Infection

Common clinical scenarios where this could be considered in the setting of contingency capacity include:

1. Staphylococcus aureus bacteremia (SAB): 2 sets of BCx are recommended due to improved sensitivity [9]. There are limited data evaluating whether 1 set of BCx is adequate to document clearance of SAB. In 1 retrospective study, the negative predictive value for excluding intermittent bacteremia was 87% for a single negative BCx on day 2, but ≥95% for single BCx >4 days [42]. Therefore, deviations from the recommended best practice of 2 sets need to weigh the potential impact of missing a positive culture (eg, how the BCx result may influence a surgeon's decision to operate for source control, how it may impact duration of therapy choices). When a patient has had >4 days of bacteremia or is already committed to a prolonged duration of therapy, spacing repeat BCx to every 3–4 days and single sets could be considered during contingency capacity, allowing 2 sets for those not meeting these criteria.

2. Blood culture contamination (BCC): While repeat BCx are not recommended for immunocompetent patients with low risk of endovascular infection, there are situations in which the significance of coagulase-negative staphylococci may be uncertain (immunocompromised patient, presence of prosthetic material). A single set could be considered for these cases during contingency capacity.

3. Enterococcus faecalis bacteremia: E faecalis can often be a contaminant in a single bottle; however, 10% of E faecalis bacteremias are associated with endocarditis [43]. Consider using the DENOVA score (Duration of symptoms, Embolization, Number of positive cultures, Origin of infection unknown, Valve disease, Ausculatation of murmur) in E faecalis bacteremia to determine the risk for endocarditis rather than repeating BCx for all patients (ie, repeat BCx if DENOVA score is high), as a DENOVA score ≥3 was shown to be 100% sensitive and 85% specific for infective endocarditis [44].

CRISIS CAPACITY

This includes strategies that deviate from best practices/standard of care and are temporarily implemented due to critical shortages to preserve the capacity to obtain BCx. Below, we discuss some strategies that would fall into this category.

Initial Single Blood Culture Sets

While restricting all BCx orders to single sets will rapidly reduce BCx utilization, it must be considered a last resort intervention used only when conventional and contingency capacity strategies were insufficient to address limited BCx bottle supplies [9] due to reduced sensitivity of single sets. Single BCx may identify 73% of BSIs versus 88%–90% with 2 sets and 97%–98% with 3 sets [9]. The yield of single BCx is influenced by species (eg, a single set may detect approximately 90% of S aureus BSIs but only 60% of Pseudomonas aeruginosa and Candida albicans cases) [9]. If this strategy is implemented, allowances for 2 sets of BCx should be considered for patients at highest risk such as those with septic shock or endovascular infections.

Use of Anaerobic Bottles Only for Selected Blood Cultures

Many bacterial organisms grow as well, or better, in anaerobic bottles than aerobic bottles [45]. When an organism known to grow well in anaerobic bottles (eg, S aureus) has been detected and if anaerobic bottles are in greater supply than aerobic bottles, repeat BCx to document organism clearance could be performed using only anaerobic bottles. Exclusive use of anaerobic bottles for initial evaluation will reduce sensitivity, as there are several common pathogens (eg, P aeruginosa, Acinetobacter, C albicans) that require aerobic bottles for detection [45].

Use of Fungal/Mycobacterial or Pediatric Bottles for Adults

Use of BCx bottles not designed for adult BCx is problematic, due to both volume limitations (multiple bottles required to achieve recommended adult blood volume) and media not validated for routine bacterial pathogen detection. It should be considered only if no other options exist and in consultation with the laboratory director.

Measuring Impact of Interventions, Including Unintended Adverse Effects

Foundational to managing any shortage is the monitoring of both utilization and potential patient impact. As contingency and crisis strategies deviate from standard care, hospitals should strive to monitor some safety metrics. We outline possible measures to monitor:

1. BCx positivity: BCx utilization or positivity benchmarks have not been established, which makes defining appropriate utilization levels challenging. Measuring BCx positivity may be a surrogate for appropriateness of BCx use (eg, if baseline utilization positivity is 10% and drops to 1% with lower BCx use, this may be a signal of missed bacteremias). When measuring positivity, BCC should be excluded (ie, only true positivity is measured).

2. BCx appropriateness/audits: While this provides more useful information to influence practices, it requires detailed chart review. Woods-Hill et al used a standardized form to assess whether a BCx stewardship program for critically ill children without sepsis may have caused delays in bacteremia detection [17]. Random sampling or targeting high-volume/low-positivity units may be considered to decrease workload. Audits with feedback should be considered before implementing more restrictive interventions such as single sets.

3. SEP-1 failure reviews: All hospitals review compliance with Centers for Medicare and Medicaid Services’ SEP-1, creating an opportunity to monitor compliance specifically with the BCx component of the bundle.

4. Monitoring of safety reporting systems: Perform vigilant monitoring of anonymous safety reports related to missed BSIs or other infectious complications due to no BCx obtainment.

CONCLUSIONS

Several studies have shown that decision support tools can assist clinicians with clinical reasoning regarding appropriate indications for BCx. These tools can be implemented safely in a variety of patient populations and should be utilized during conventional capacity. How these strategies are implemented will depend on local resources and time pressure for implementation.

Contingency and crisis strategies can be considered in the setting of a shortage but should be de-implemented when supply improves. While the BCx shortage has the potential to negatively impact patient care, this situation also presents an opportunity to adopt BCx stewardship interventions more widely, to advocate for better funding of diagnostic/antimicrobial stewardship research, and to reevaluate existing regulatory rules that conflict with diagnostic stewardship principles.

Notes

Author contributions. All authors were involved in the conception and writing of the manuscript and approved the final submitted version.

Disclaimer. The content is solely the responsibility of the authors and does not represent the official view of the funding agency.

Financial support. J. H. R., T. C. V. S., D. J. D., and V. F. receive support through the Johns Hopkins Prevention Epicenter Blood Culture Stewardship Collaborative funded by the Prevention Epicenters Program, Centers for Disease Control and Prevention (grant numbers 51U54CK000617–02-00 and 5U54CK00617-03-001).