Hostname: page-component-586b7cd67f-t8hqh Total loading time: 0 Render date: 2024-11-24T19:24:21.407Z Has data issue: false hasContentIssue false

To wait or not to wait: Optimal time interval between the first and second blood-culture sets to maximize blood-culture yield

Published online by Cambridge University Press:  25 March 2022

Valeria Fabre*
Affiliation:
Division of Infectious Diseases, Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland
George F. Jones
Affiliation:
Division of Infectious Diseases, Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland
Yea-Jen Hsu
Affiliation:
Department of Health Policy and Management, Johns Hopkins Bloomberg of School of Public Health, Baltimore, Maryland
Karen C. Carroll
Affiliation:
Division of Clinical Microbiology, Department of Pathology, Johns Hopkins University School of Medicine, Baltimore, Maryland
Sara E. Cosgrove
Affiliation:
Division of Infectious Diseases, Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland
*
Author for correspondence: Valeria Fabre, MD, Division of Infectious Diseases, Department of Medicine, Johns Hopkins University School of Medicine, 600 North Wolfe Street, Osler 421 Baltimore, MD 21287. E-mail: [email protected]

Abstract

The optimal timing of blood culture (BCx) sets collection has not been evaluated with continuous BCx detection systems. The yield of BCx was similar between short intervals (median, 3 minutes) and longer intervals (median, 16 or 43 minutes) among 5,856 BCx, except for improved polymicrobial bacteremia detection with long-interval BCx.

Type
Concise Communication
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution and reproduction, provided the original article is properly cited.
Copyright
© The Author(s), 2022. Published by Cambridge University Press on behalf of The Society for Healthcare Epidemiology of America

Blood culture (BCx) stewardship is important to ensure appropriate diagnosis of bloodstream infections, and it includes optimization of BCx collection. Reference Fabre, Carroll and Cosgrove1 Although studies have addressed the impact detection of the timing from BCx collection to incubation or in relation to fever on bacteremia, Reference Nannan Panday, Wang, van de Ven, Hekker, Alam and Nanayakkara2Reference Venturelli, Righi and Borsari4 data on the optimal interval between successive BCx sets are very limited. Historically, a wait time between BCx sets has been recommended out of concern that simultaneous BCx may miss intermitent bacteremias or bacteremias with low bacterial concentration. Reference Dean and Lee5,Reference Weinstein6 However, a study evaluating 7,783 BCx processed with the BACTEC 660 infrared detection system in 1994 showed that BCx positivity was not increased by lengthening intervals between BCx in a 24-hour period. Reference Li, Plorde and Carlson7 The American Society for Microbiology and the Clinical and Laboratory Standards Institute recommend collecting multiple sets simultaneously or over a short time, except when documentation of continuous bacteremia is required for patients with endovascular infection. Reference Baron and Baron8 The Infectious Diseases Society of America guidelines recommend sequential BCx over minutes in urgent situations when prompt antibiotic initiation is needed (eg, septic shock) and longer intervals (several hours or more) for nonurgent situations. A reference for this recommendation is not provided. Reference Miller, Binnicker and Campbell9 Furthermore, whether simultaneous BCx are widely accepted in clinical practice remains unlcear.

We evaluated the impact of timing on BCx yield in the era of continuous BCx detection instruments.

Methods

Study population

We retrospectively evaluated the first 2 sets of BCx obtained within the first 24 hours from patients ≥18 years of age presenting to the emergency department at The Johns Hopkins Hospital (JHH) between January 28, 2017, and October 31, 2019. JHH is an academic hospital in Baltimore, Maryland. We excluded patients with additional BCx obtained beyond the first 2 sets to avoid bias associated with increased likelihood of bacteremia detection with additional sets, Reference Lee, Mirrett, Reller and Weinstein10 solitary BCx (ie, 1 BCx set in a 24-hour period), and BCx growing contaminants.

BCx were processed in the JHH Microbiology Clinical Laboratory using the BD Bactec FX BCx system (Becton Dickinson, Sparks, MD) for organism detection. BCx bottles were incubated for 5 days, and identification of microorganisms was performed using matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) and the Verigene BCID-GP panel. BCx collection times and results were electronically extracted from the electronic health record (EHR). The recommended volume for BCx is 10 mL per bottle for adults.

Definitions

BCx contamination was defined using a standard laboratory definition (growth in a single BCx set of any of the following organisms: Bacillus spp, coagulase-negative staphylococci, Corynebacterium spp, Micrococcus spp, Cutibacterium acnes, and viridans group streptococci). A BCx pair refers to the first 2 sets of BCxs drawn from a patient (1 set includes 1 aerobic and 1 anaerobic bottle). BCx positivity was defined as the proportion of BCx pairs with growth of a noncontaminant organism in 1 or both BCx sets. BCx with collection times in Epic software (Epic, Verona, WI) separated by 0–9 minutes were defined as short-interval or simultaneous blood cultures, and sets separated by ≥10 minutes (10–29 minutes or ≥30 minutes) were defined as long-interval BCx.

Statistical analysis

BCx positivity was compared across intervals and were stratified by causative organism. Analyses utilized χ2 tests or Fisher exact tests when appropriate (first P value reported). Nonparametric tests for trend were used to further evaluate the impact of periodicity on BCx positivity. We accounted for potential differences in BCx practices in time through a regression analysis. A 2-sided P value <.05 was considered statistically significant. Analyses were performed using Stata software (2019, StataCorp, College Station, TX).

This study was approved by the Johns Hopkins Medicine Institutional Review Board.

Results

Of 6,938 BCx identified within the study period, 1,082 (15.6%) were excluded: 857 due to single BCx sets, 161 due to ≥3 BCx sets in 24 hours, and 64 due to BCx contaminants. Of the 5,856 BCx sets included in the analysis, or 2,928 pairs, 789 (27%) of 2,928 pairs were drawn 0–9 minutes apart (median, 3 minutes; interquartile range [IQR], 0–8); 1,733 (59%) were drawn 10–29 minutes apart (median, 16 minutes; IQR, 13–21); and 406 (14%) were drawn with a ≥30-minute interval (median 43 minutes; IQR, 34–80). The proportion of BCx drawn 0–9 minutes increased from 23.5% in 2017 to 36% in 2018 to 40% in 2019 (P < .01 for all comparisons).

The overall BCx positivity rate was 15% (439 of 2,928); most cases were monomicrobial, and 57 (13%) of these 439 cases were polymicrobial. BCx positivity rates were similar among the 3 collection time groups after adjusting for month or year of BCx collection: 13% (105 of 789) for 0–9 minutes, 15% (265 of 1,733) for 10–29 minutes, and 17% (69 of 406) for ≥30 minutes (P = .20, test for trend P = .07) (Table 1) (Supplementary Material). In a subgroup analysis excluding 274 BCx pairs with antibiotics given before the first or second BCx set, rates remained similar (14%, 16%, and 15% positivity for the 3 groups, respectively; P = .40; test for trend P = .27) (Supplementary Material).

Table 1. Blood Culture (BCx) Positivity by Time Interval and by Causative Pathogen

a First P values are from χ2 or Fisher exact test; second P values are from nonparametric tests for trend.

We detected no differences in BCx positivity among short- or longer-interval BCx that grew gram-positive organisms, gram-negative organisms, or anaerobes, although we detected a non-significant trend toward increased yield for anaerobes among BCx obtained ≥30 minutes apart: 0.6% for 0–9 minutes, 0.7% for 10–29 minutes and 1.7% for ≥30 minutes (P = .12; test for trend P = .09). We also detected a trend toward increased yield for polymicrobial bacteremias for longer-interval BCx: 10 (9.5%) of 105 for 0–9 minutes; 33 (12%) of 265 for 10–29 minutes; and 14 (20%) of 69 for ≥30 minutes (P = .10; test for trend P = .04). For yeast, we also detected a trend toward increased yield: 0 (0%) for 0–9 minutes; 4 (0.2%) of 1,733 for 10–29 minutes; and 3 (0.7%) of 406 for ≥30 minutes (P = .04; test for trend P = .01).

Discussion

We investigated the impact of time between the first and second BCx sets on BCx yield among 5,856 BCx from adult patients using a continuous BCx detection system. We found similar yield for common bacterial pathogens when sets were collected within a short interval (0–9 minutes) or a longer interval (≥10 minutes).

This study had several limitations. The interval between BCx sets derived from the EHR depends on when the individual drawing the BCx scanned the specimen into the EHR. However, we did not expect the records to be biased in one direction or the other, and simultaneous BCx are acceptable at our institution. We did not have data on blood volume per individual bottle, which may have affected bacteremia detection, because the latter is measured in aggregate using instrument software. However, we included a large number of BCx, which would have mitigated any potential disparities in blood volume in the different time groups. Also, we adjusted for potential differences in BCx practices in time through regression analysis. The importance of timing between BCx sets for polymicrobial and yeast bloodstream infections warrants further investigation in larger studies. Finally, difficult-to-grow organisms were not well represented in our cohort, and additional study is warranted to understand the impact of timing in these scenarios.

In summary, our data indicate that the timing of the second BCx set does not impact the yield of BCx to detect most common bacterial pathogens. Collecting BCx simultaneously may help streamline the clinical workflow. Future larger studies including source of infection or patient characteristics could help determine whether longer intervals improve detection of polymicrobial and yeast bloodstream infections and which patients would benefit from this practice.

Supplementary material

To view supplementary material for this article, please visit https://doi.org/10.1017/ash.2022.27

Acknowledgments

The authors thank Eili Klein, PhD, for assisting with data extraction.

Financial support

This study was funded in part by the Centers for Disease Control and Prevention’s Prevention Epicenters Program (grant no. 1 UK54CK000617-01-00).

Conflicts of interest

All authors report no conflicts of interest relevant to this article.

References

Fabre, V, Carroll, KC, Cosgrove, SE. Blood culture utilization in the hospital setting: a call for diagnostic stewardship. J Clin Microbiol 2021. doi: 10.1128/JCM.01005-21.Google Scholar
Nannan Panday, RS, Wang, S, van de Ven, PM, Hekker, TAM, Alam, N, Nanayakkara, PWB. Evaluation of blood culture epidemiology and efficiency in a large European teaching hospital. PLoS One 2019;14:e0214052.CrossRefGoogle Scholar
Riedel, S, Bourbeau, P, Swartz, B, et al. Timing of specimen collection for blood cultures from febrile patients with bacteremia. J Clin Microbiol 2008;46:13811385.CrossRefGoogle ScholarPubMed
Venturelli, C, Righi, E, Borsari, L, et al. Impact of pre-analytical time on the recovery of pathogens from blood cultures: results from a large retrospective survey. PLoS One 2017;12(1):e0169466.CrossRefGoogle ScholarPubMed
Dean, AJ, Lee, DC. Bedside laboratory and microbiologic procedures. In: Roberts and Hedges’ Clinical Procedures in Emergency Medicine and Acute Care, seventh edition. New York: Elsevier; 2017:1442–1469.Google Scholar
Weinstein, MP. Current blood culture methods and systems: clinical concepts, technology, and interpretation of results. Clin Infect Dis 1996;23:4046.CrossRefGoogle ScholarPubMed
Li, J, Plorde, JJ, Carlson, LG. Effects of volume and periodicity on blood cultures. J Clin Microbiol 1994;32:28292831.CrossRefGoogle ScholarPubMed
Baron, EJ, et al. Cumitech 1C, Blood Cultures. Coordinating editor, Baron, EJ. Washington, DC: ASM Press; 2005.Google Scholar
Miller, JM, Binnicker, MJ, Campbell, S, et al. A guide to utilization of the microbiology laboratory for diagnosis of infectious diseases: 2018 update by the Infectious Diseases Society of America and the American Society for Microbiology. Clin Infect Dis 2018;67(6):e1e94.CrossRefGoogle Scholar
Lee, A, Mirrett, S, Reller, LB, Weinstein, MP. Detection of bloodstream infections in adults: how many blood cultures are needed? J Clin Microbiol 2007;45:35463548.CrossRefGoogle ScholarPubMed
Figure 0

Table 1. Blood Culture (BCx) Positivity by Time Interval and by Causative Pathogen

Supplementary material: File

Fabre et al. supplementary material

Fabre et al. supplementary material

Download Fabre et al. supplementary material(File)
File 18.9 KB