If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Contaminated blood cultures may have detrimental effects on patients, the organization, and antimicrobial stewardship. Patients in the emergency department may need blood cultures collected before antimicrobial therapy. Contaminated blood culture samples may contribute to prolonged hospital stay and also are associated with delayed or unnecessary antimicrobial therapy. This initiative aims to improve the emergency department’s blood culture contamination rate that will eventually benefit the patients who will receive timely and proper antimicrobial therapy, and benefit the organization fiscally.
Methods
This quality improvement initiative used the Define–Measure–Analyze–Improve–Control (DMAIC) process. The organization targets blood culture contamination rate of ≤2.5%. Control charts were used to study how blood culture contamination rate changed over time. In 2018, a workgroup was formed to work on this initiative. Improved site disinfection using 2% Chlorhexidine gluconate cloth before the standard procedure of blood culture sample collection was initiated. Chi squared test of significance was used to compare blood culture contamination rates 6 months before and during feedback intervention as well as contamination rate from source of blood draw.
Results
Blood culture contamination rates 6 months before and during feedback intervention showed significant decrease (3.52% before intervention and 2.95% after intervention; P < .05). Contamination rates differed significantly based on the source of blood culture draw (7.64% via line, 3.05% via percutaneous venipuncture, and 4.53% via other; P < .01).
Discussion
Blood culture contamination rate continued to decrease with the use of a predisinfection process with 2% Chlorhexidine gluconate cloth before blood sample collection process. Practice improvement also was evident with effective feedback mechanism.
Current literature supports the need to reduce blood culture contamination rates in emergency departments. However, limited evidence is found investigating the comparison of blood culture contamination rates collected from percutaneous venipuncture to other collection sites.
•
This article contributes evidence on the effectiveness of specific evidence-based interventions to reduce blood culture contamination rates, including predisinfection with 2% Chlorhexidine gluconate cloths when collecting samples for blood cultures, and surveillance and feedback.
•
Key implications for emergency nursing practice include integrating evidence-based practice changes with a robust feedback mechanism to reduce blood culture contamination rates in the emergency department setting.
Introduction
Blood cultures are important diagnostic tools for identifying the pathogens responsible for a patient’s infection.
Contaminated blood cultures may have detrimental effects (eg, unnecessary antibiotic exposure and prolonged length of stay) on the patient, to the organization, and to antimicrobial stewardship efforts.
Other consequences include unnecessary antibiotic exposure with the potential for downstream unintended consequences (eg, possible allergic reactions and Clostridioides difficile infection).
Practical guidance for clinical microbiology laboratories: a comprehensive update on the problem of blood culture contamination and a discussion of methods for addressing the problem.
Estimated clinical and economic impact through use of a novel blood collection device to reduce blood culture contamination in the emergency department: a cost-benefit analysis.
found that the average length of stay was 2 days longer in patients with contaminated blood cultures than in patients with negative cultures. That same study found that direct and indirect hospital costs of a contaminated blood culture were $12,824 compared with $8286 for a negative blood culture (cost savings of $4538 for preventing a contaminated blood culture).
Estimated clinical and economic impact through use of a novel blood collection device to reduce blood culture contamination in the emergency department: a cost-benefit analysis.
found that blood culture contamination was more likely to occur in critically ill patients, that is, triage levels 1 and 2 (modified Canadian Triage and Acuity Scale), probably because these patients received urgent care, restricting the time for appropriate blood sampling procedures. The same study
adds that underlying conditions, that is, end-stage renal disease and older age, were associated with blood culture contamination in emergency departments. Chang et al
also discussed that these patients might frequently visit health care facilities and potentially carry skin commensals with antimicrobial resistant genes. In addition, it was explained that blood draw challenges were more likely in these patient groups due to poorly accessible veins.
Literature shows several strategies taken by different organizations to improve their blood culture contamination rates. Self et al
Practical guidance for clinical microbiology laboratories: a comprehensive update on the problem of blood culture contamination and a discussion of methods for addressing the problem.
found that the use of blood culture collection kits and standardized procedures have been associated with a significant decrease in blood culture contamination. Surveillance and feedback systems also have been shown to result in improved blood culture contamination rates, particularly when contamination rates are reported in a timely manner and directed individually to those who collected the samples.
The University of California Davis Medical Center emergency department’s blood culture contamination rates were noted to be above the target rate of 2.5%, ranging from 2.79% to 7.28% from 2018 to 2020. The department runs approximately 40 sets of blood culture tests per day. Reducing blood culture contamination aligns with the emergency department’s strategic quality and financial stewardship goals. A multidisciplinary workgroup worked robustly to improve this metric using evidence-supported practice changes.
Methods
This process improvement initiative, which was started in 2018 by University of California Davis Medical Center emergency department, used the Define–Measure–Analyze–Improve–Control (DMAIC) model of quality improvement. Data (blood culture contamination rates, dates and times of sample collection, and patients’ health record numbers) were collected from EPIC (Epic Systems Corporation) electronic health record reports that feed daily to the organizational dashboards.
This initiative was rolled out in 2018 and has been an ongoing process of analyzing issues, addressing identified issues, and evaluating measures undertaken. Spot analyses were performed as needed. Descriptive statistics were used to study specific time frames (spot analyses for July 2020 and August to September 2021). Contamination rates were plotted against control charts by intervention phases to determine process control and special cause variation. Contamination rates 6 months before and during feedback intervention and contamination rate difference based on source were analyzed using Chi squared test of significance. Statistical analyses were performed using Stata 17.0 (StataCorp LLC). Reporting guidelines
Blood culture contamination: an overview for infection control and antibiotic stewardship programs working with the clinical laboratory. Centers for Disease Control and Prevention.
states that because blood is a normally sterile body site, positive blood cultures with a known pathogen have a generally overall high positive predictive value for infection. However, blood culture contamination is a significant problem. In general, all blood culture contamination occurs during collection; the source of contaminants is usually the patient’s skin or the hub or cannula of an indwelling catheter (ie, when an existing catheter is used to obtain the specimen). Frequent causes include poor collection technique and insufficient skin disinfection.
Blood culture contamination: an overview for infection control and antibiotic stewardship programs working with the clinical laboratory. Centers for Disease Control and Prevention.
The organization targets a blood culture contamination rate of ≤2.5%. The emergency department’s blood culture contamination rate from 2018 to 2020 ranged from 2.79% to 7.28% (fiscal year 2019-2020, mean = 4.56%; fiscal year 2020-2021 mean = 3.86%).
The University of California Davis Medical Center determines blood culture contamination by the number of contaminating organisms from percutaneous and/or line blood draws per total blood culture samples collected. If blood culture samples yield to growth of an organism that is not a true pathogen or when multiple nonpathogenic organisms are identified, it will be flagged contaminated (Jordan Jones, CLS, email communication, August 08, 2022).
Whenever an organism on the list is identified, the electronic health record automatically flags the cultures as a possible contamination. Contamination flags and comments are not removed when a provider requests susceptibility testing unless there is >50% of blood culture sets, for an episode (3-day collection period), turned positive with 1 morphotype, and there are no other organisms present in the cultures. There also must be more than one set of culture bottles and this generally applies to the coagulase-negative Staphylococci and Streptococcus species. Contamination flags also are not routinely removed for other organisms on the contamination list. Blood culture samples from central venous access devices (CVADs) and percutaneous venipuncture collection sets are not treated differently—determining whether there are >50% of the sets collected within a 3-day episode positive with the same single organism. For example, if there are multiple coagulase-negative Staphylococci identified in a report denoting multiple morphotypes, the contamination flag will not be removed if one of the 2 organisms is present in multiple sets.
Analyze
Through structured problem-solving and continuous improvement approach, several causes of sample contamination were identified as follows: education gap, inappropriately prepared venipuncture site, supplies needed for sample collection were not stored in one location, blood cultures were not included in initial workup orders, fast-paced workflows, time-sensitive procedures, variety of patient population (prehospital environment significantly adds to skin contaminants), drawing blood culture samples from existing intravenous access, contaminated samples drawn from ultrasound-guided intravenous access insertion, contaminated samples from CVADs, and contaminated samples among pediatric patients, patients with coronavirus, and critically ill patients.
Improve
A multidisciplinary workgroup (consisting of ED leadership, clinical nurse leaders, clinical resource nurses, clinical nurses, infection prevention, quality and safety, laboratory) worked collaboratively to improve this metric since 2018. The team worked to heighten department awareness on its standing on this metric. Information was continuously shared through preshift huddles and workgroup meetings wherein strong team engagement was present. Education was reinforced through several avenues such as incorporating content in new-hire orientation, just-in-time coaching, and periodic skills day. Blood culture sample collection has been a standardized procedure for nurses and was added to the Best Practice Advisory for patients meeting Sepsis Analytic Model score ≥8. The use of diversion devices, which passively sideline skin contaminants, had been considered but was not voted on due to cost ($15 per device); alternatively, the team decided to focus on reinforced skin antisepsis instead (Table 1). Notably 2% Chlorhexidine gluconate cloths were added (cost $6) to the standard blood culture sample collection supplies with the purpose of thoroughly cleansing the skin surface before blood draw. The added step of cleaning the site with 2% Chlorhexidine gluconate cloth was not applicable for patients sensitive to chlorhexidine or patients younger than 2 months of age.
Because of limited safety data, chlorhexidine is not recommended for use in children <2 months of age.11
Not applicable for patients with hypersensitivity or patients <2 mo of age.
2.
Perform hand hygiene.
3.
Clean the blood culture bottle tops.
a.
Cleanse tops with 3.15% Chlorhexidine gluconate/70% Isopropyl alcohol pad for 5 s and dry for 5 s or 70% Isopropyl alcohol pad for 15-30 s.
b.
Allow to dry.
4.
Scrub sample collection site in a back-and-forth motion.
a.
Clean the skin with 70% Isopropyl alcohol pad for 15-30 s. Allow to dry.
b.
Disinfect the skin with 2% Chlorhexidine gluconate/70% Isopropyl alcohol applicator for 5 s. Allow to dry. For patients who are sensitive to chlorhexidine or patients <2 mo of age, cleanse the skin using 10% Povidone-iodine sticks. Allow to dry.
∗ Because of limited safety data, chlorhexidine is not recommended for use in children <2 months of age.
It also was emphasized not to draw blood culture samples from existing intravenous accesses. For CVADs, education was rolled out not to routinely draw from these accesses unless the provider suspects infection from the source and orders collection from the access; in that case, one set of blood cultures need to be collected from the CVAD after scrubbing the hub, changing the needleless connector, and scrubbing the hub again, without discarding the first blood draw,
and the other set has to be collected from another source. The new process was reviewed by stakeholders and was finalized and published as a departmental policy. To reduce variation in practice, an instructional video demonstrating the practice change was developed and socialized to all ED staff, new-hire and temporary staff included. Dedicated phlebotomists were hired to support staffing-related issues.
Control
ED blood culture contamination rate has been closely monitored since 2018. Starting with overall departmental rates, data available were optimized, generating inferences such as contamination rate per individual collector and contamination by microorganisms. Chart reviews were conducted to determine rates by patient acuity and room placement, time of collection, and other circumstances surrounding contamination. Monthly organizational recognition was given to staff nurses who had the greatest number of blood cultures collected without contamination. Regular just-in-time feedback was provided to staff nurses who collected contaminated blood cultures, allowing discussion to identify contributing factors and risk mitigation. Nurses who had repeated patterns or increased contamination rate without improvement were referred to department leadership for further action (ie, repeat skills check-off with clinical resource nurse or educator). Individual contamination rate also was added as a discussion point in the staff performance evaluation wherein department leadership had the opportunity to revisit this skill with staff.
Findings
From January to December 2018, the workgroup was formed. Possible causes of blood culture contamination were identified. Teaching and coaching were found not effective. This prompted the trial of adding predisinfection with 2% Chlorhexidine gluconate cloth before blood culture sample collection process. Trends could not be drawn from blood culture contamination rates; unpredictable upticks and decline were noted. By December 2018, the contamination rate was at its highest at 7.28%.
In 2019, reports also were validated. The use of a diversion device in collecting blood culture samples was considered but was held due to cost. At that time, the contamination rate started to show less variation as demonstrated by narrower control limits. Contamination rate during this period ranged from 3.36% to 5.72%.
In 2020, contamination rate fluctuation was noted as well as increased practice variation, which is indicated by widening control limits. A spot analysis done in July 2020 shows that 58% of contaminated blood culture samples were from critically ill patients with shortness of breath as their chief complaint and that the median time from room to blood culture draw was 36 minutes. Discussion was started on possibly adding blood culture collection order on critically ill patients or on some targeted patient population. Dedicated phlebotomists also was started but later met staffing challenges. Blood culture contamination rate during this period ranged from 3.56% to 6.44%.
From January to August 2021, it was determined that the predisinfection with 2% Chlorhexidine gluconate cloths was helping reduce blood culture contamination, and a request was made to produce blood culture collection kits with this additional product. A Best Practice Advisory to order blood cultures also was added for patients whose Sepsis Analytic Model score was high (≥8). Reports received were further analyzed, and contamination rate per collector was added to the reports enabling more targeted feedback. Decrease in contamination rate and narrowing of practice variation were notable during this period. Blood culture contamination rate ranged from 2.20% to 4.47%. The target blood culture contamination rate was met by the department on 3 occasions: February, May, and June 2021.
The following details were drawn from data collected from August to September 2021: 85.7% of contaminated blood samples (N = 91) were collected from the main emergency department (adult pods and resuscitation area), 8.79% were collected from boarding ED patients, 3.3% were collected from pediatric patients, and 2.2% were collected from rapid care area. All contaminated samples (N = 91) from this period were collected within 36 hours of patient arrival to emergency department, 10% of contaminated samples were taken within the first 30 minutes of patient’s ED arrival, 25% of contaminated samples were extracted within the first 65 minutes of patient’s ED stay, 50% of contaminated samples were taken within the first 3 hours of ED stay, and 75% were taken within the first 7 hours of patient’s ED stay.
Another spot analysis showed that there were 137 contaminated blood culture samples from July to September 2021, 25% of which were collected from 12 midnight to 09:59 AM, 25% were taken from 10 AM to 02:59 PM, 25% were taken from 3 PM to 07:59 PM, and the other 25% were collected from 8 PM to 11:59 PM. Thus, 50% of the contaminated samples were collected from 12 AM to 02:59 PM, and the other half were collected from 3 PM to 11:59 PM.
Control measures were reinforced since September 2021. On a daily basis, collectors whose samples turned contaminated were notified by email. This process prompted in-person discussion and feedback. Staff nurses were receptive to this process, which is timely, collegial, constructive, and individualized. To further decrease practice variation, a video demonstrating how to use the blood culture collection kit was developed and shared with all staff nurses. This content also was included in the new-hire staff nurses’ orientation program. Reward and escalation workflows also were established. Contamination rates of staff nurses also have been included as a discussion point in their performance evaluation. This period has shown a progressive decrease in variation and a downward trend in contamination rate. Blood culture contamination rate in this period ranged from 2.86% to 4.64%.
Blood culture contamination rates also were compared 6 months before (March to August 2021) and 6 months during (September 2021 to February 2022) feedback intervention and showed significant decrease (3.52% preintervention and 2.95% postintervention; P < .05) (Table 2). From March 2021 to February 2022, blood culture contamination rates significantly differed based on the source of blood culture draw (7.64% via line, 3.05% via percutaneous venipuncture, and 4.53% via other; P < .01) (Table 3).
Table 2Test of significance: contamination rate before and during feedback intervention
Contaminated
6 mo during feedback (March-August 2021)
6 mo before feedback (September 2021-February 2022)
Figure 1 shows the emergency department’s blood culture contamination rate from January 2018 to July 2022, which indicates that the overall monthly rate shows a downward trend. From September 2021 to July 2022 (Figure 2), contamination rate from blood culture samples collected from percutaneous venipuncture ranged from 2.27% to 4.39%. Alternatively, from September 2021 to July 2022, contamination rate from blood culture samples collected from other sources (eg, peripherally inserted central catheter, CVADs) (Figure 3) ranged from 2.17% to 8.16%. Figure 4 demonstrates that the emergency department’s monthly overall contamination rates were strongly attributable to the contamination rates from other sources. For instance, in July 2022, the overall contamination rate was 3.06%. It can be noted that contamination rate from percutaneous venipuncture was only 2.67%, but this was greatly affected by the contamination from other sources, which was 6.67%.
Figure 1Blood culture contamination rate in ED from January 2018 to July 2022. ED, emergency department; SAM, Sepsis Analytic Model.
Reducing blood culture contamination in the emergency department has been a challenge due to several factors such as patient volume and acuity, fast-paced workflows, time-sensitive procedures, and staff transition. This improvement initiative investigated and integrated effective measures (ie, use of dedicated phlebotomy,
Multidisciplinary team review of best practices for collection and handling of blood cultures to determine effective interventions for increasing the yield of true-positive bacteremias, reducing contamination, and eliminating false-positive central line-associated bloodstream infections.
Multidisciplinary team review of best practices for collection and handling of blood cultures to determine effective interventions for increasing the yield of true-positive bacteremias, reducing contamination, and eliminating false-positive central line-associated bloodstream infections.
Site predisinfection with 2% Chlorhexidine gluconate cloths significantly reduced skin contaminants. It can be noted that ED patients come from various circumstances such as private residences, the field, or other facilities. This alone is an uncontrollable variable in maintaining aseptic technique during specimen collection indicating the importance to adhere to aseptic technique. Staff nurses were remarkably found to be more receptive and engaged in blood culture contamination reduction initiative with timely, informal, collegial, and individualized feedback.
It remains a challenge to reduce contamination from blood culture samples collected from other sources, such as CVADs. In collecting such samples, the emergency department adheres to scrubbing the hub, changing the needleless connector, and scrubbing the hub again, drawing blood culture samples without discarding the first blood draw.
It is of great interest how blood culture samples collected from these sources turned contaminated in the emergency department despite compliance to established workflow. There were several instances wherein blood culture samples collected from a CVAD resulted as contaminated in the emergency department, and upon recollection on the floor (for an instance, repeat blood culture within 3 days of admission), blood culture samples collected from the same CVAD was not flagged contaminated.
Conclusion and Recommendations
In this 4-year period, the emergency department only reached the target rate on 3 occurrences. Keeping the contamination rate ≤2.5% has been a challenge, but the department is making progress in gradually reducing it and narrowing practice variation. Key drivers for the success of this practice and process improvement initiative are reinforced skin antisepsis (predisinfecting the site with 2% Chlorhexidine gluconate cloth before blood culture sample collection) and an effective surveillance and feedback mechanism. It is critical to note that predisinfecting the sample collection site was instrumental in this initiative. Of equal importance is leveraging timely and individualized feedback to sample collectors.
To date, the department has gained control in reducing contamination on blood cultures collected from percutaneous venipuncture. Future direction is headed to sustaining progress, reinforced coaching and feedback, and more focus on other sources (ie, CVADs) given that the department’s overall contamination rate is significantly affected by the increased contamination from these sources.
The use of diversion device might be worth reconsidering given that it was found effective in some studies.
Multidisciplinary team review of best practices for collection and handling of blood cultures to determine effective interventions for increasing the yield of true-positive bacteremias, reducing contamination, and eliminating false-positive central line-associated bloodstream infections.
The authors will continue to search, appraise, and synthesize evidence on (1) how skin contaminants differ from contaminants found on lines and (2) how to further reduce blood culture contamination, more specifically in samples collected from other sources.
Author Disclosures
Conflicts of interest: None to report.
Acknowledgments
The authors thank Toby Marsh, MSA, MSN, RN, FACHE, NEA-BC; Stacy Hevener, MSN, RN, CSSGB, CPHQ; Brynne Kessler, MSN, RN-BC, CSSGB, CPHQ, and University of California Davis Medical Center PCS Quality and Safety Department; Rupinder Sandhu, MBA, RN; Frances Noriega, MSN, RN; Jaime Heitmeyer, MBA, RN, CEN; Swapna Peter, MSN, RN, CCRN; University of California Davis Medical Center ED Leadership; UC Davis Medical Center ED Unit-Based Practice Council; Yvonne Hansen, MS, RN, CCRN, CEN, TCRN; Marsha Hoeft, BSN, RN, CEN, MICN, TCRN; Tara Lynn Barragan, BSN, RN, CEN, TCRN, CPEN; Keith Taggatz, BSN, RN, TNCC, MICN; Anika Kutschmar, BSN, RN, CEN; ED Clinical Nurse Leaders, Clinical Resource Nurses and Clinical Nurses; Sarina Fazio, PhD, RN; and Lori Kennedy, PhD, RN, ACNP-BC.
References
Rhodes A.
Evans L.E.
Alhazzani W.
et al.
Surviving sepsis campaign: international guidelines for management of sepsis and septic shock: 2016.
Practical guidance for clinical microbiology laboratories: a comprehensive update on the problem of blood culture contamination and a discussion of methods for addressing the problem.
Estimated clinical and economic impact through use of a novel blood collection device to reduce blood culture contamination in the emergency department: a cost-benefit analysis.
Blood culture contamination: an overview for infection control and antibiotic stewardship programs working with the clinical laboratory. Centers for Disease Control and Prevention.
Multidisciplinary team review of best practices for collection and handling of blood cultures to determine effective interventions for increasing the yield of true-positive bacteremias, reducing contamination, and eliminating false-positive central line-associated bloodstream infections.
Charlotte Marcelino is a Quality and Safety Champion with Patient Care Services, Quality and Safety Department in University of California Davis Medical Center, Sacramento, CA. ORCID identifier:https://orcid.org/0000-0002-4318-5944.
Jan Shepard is a Nursing Professional Development Specialist with Center for Professional Practice of Nursing in University of California Davis Medical Center, Sacramento, CA. ORCID identifier:https://orcid.org/0000-0003-0215-3003.
00309-9&id=gr1.jpg){kind=link}
00309-9&id=gr2.jpg){kind=link}
00309-9&id=gr3.jpg){kind=link}
00309-9&id=gr4.jpg){kind=link}