Bringing doctors together

or

A free, private network for verified physicians

Leading voices

Dr. Christopher Graber

Antibiotic Stewardship · Greater Los Angeles VA

View

Dr. Jonathan Ryder

Antibiotic Stewardship · UNMC

View

Dr. Chris Beyrer

Global Health · Duke University

View

Dr. Lance Okeke

HIV/AIDS Medicine · Duke University

View

Dr. Raphael J. Landovitz

HIV/AIDS Medicine · UCLA

View

Dr. Keith Kaye

General ID · RU

View

Dr. Helen Boucher

Transplant/Immunocompromised States · Tufts

View

Dr. Sara Dong

Transplant/Immunocompromised States · Boston Children's Hospital

View

Dr. Aniruddha Hazra

HIV/AIDS Medicine · UCMC

View

Dr. Rochelle Walensky

General ID · Harvard Medical School

View
  • Discover your community

  • Stay current

  • A profile that represents you

  • Find your people

About RoonPrivacyTerms

    Overview

    Comparing contamination rates in paired blood culture sets with a single-time stamp vs multiple-time stamps.

    Sep 16, 2025DOI: 10.1017/ash.2025.10103Journal ArticleASHE

    Written by: Abhilasha BoradAli IbrahimSusan E BoruchoffKeith S KayeAhmed Abdul Azim

    Find these authors on Roon

    AA
    Dr. Ahmed Abdul Azim
    Antibiotic Stewardship · RU

    Introduction

    The current recommendations regarding blood culture (BC) methods specify that optimal BC sampling in adults consists of collecting 20–30 mL of blood per culture set, evenly distributed between aerobic and anaerobic bottles, with two sets (totaling 40–60 mL) obtained from separate venipuncture sites, collected sequentially.1,2 Improper sampling methods increase the risk of BC contamination, leading to unnecessary hospital admissions, inappropriate antibiotic therapy, prolonged lengths of stay, and increased healthcare costs.3,4Clinicians at our academic tertiary center, Robert Wood Johnson University Hospital, noticed repeated instances where two sets of BCs obtained in the Emergency Department (ED) with identical collection time stamps in the Electronic Medical Record (EMR) grew commensal organisms in both sets in patients with low pretest probability for true bacteremia. The policy at our hospital is to obtain BC paired sets from different peripheral sites when possible. We hypothesized that these instances where two BC sets had the same time stamp indicated that blood was obtained from a single stick at one anatomic site and inoculated into both sets of BC bottles, rather than from two separate sticks and anatomic sites. This assumption was supported in the ED setting based on conversations with ED staff and attendings, as well as various patients over the years who reported that four BC bottles were all drawn from a newly placed peripheral intravenous (IV) catheter.We evaluated the recovery rate of commensal organisms in both sets of BCs, predicting that rates would be higher in the single-time stamp (STS) BC group compared to the multiple-time stamps (MTS) group.

    Methods

    This is a single-center retrospective analysis using data from clinical microbiology and hospital databases. The primary study outcome was the rate of recovery of predefined commensals5 (see Table 1), which were compared for patients with two sets of BCs drawn with identical time stamps (categorized as STS) and for patients with two BC sets drawn with different time stamps within 1 hour of one another (categorized as MTS). All paired BC sets, regardless of whether cultures were obtained peripherally or from a central line, from adult patients who presented to our 650-bed tertiary care center between June 1, 2022, and November 15, 2023, were included. Pairs of BC sets with non-commensal organisms isolated in any set were excluded. Collected data includes BC timing, BC results, and admission unit. Rates from the ED and non-ED inpatient locations (which included both intensive care unit and medical-surgical wards) were also compared. We excluded patients under 18 years of age.Table 1.Predefined commensal organismsActinomyces spp.Staphylococcus auricularis*Aerococcus sanguinicolaStaphylococcus capitisAerococcus urinae spp.Coagulase-negative StaphylococcusArthrobacter spp.Staphylococcus epidermidisBacillus spp.Staphylococcus haemolyticusBrevibacterium spp.Staphylococcus hominisCorynebacterium spp.Staphylococcus pettenkoferiCutibacterium acnesStaphylococcus saprophyticusDermabacter spp.Staphylococcus schleiferiJanibacter spp.Staphylococcus simulansKocuria spp.Streptococcus anginosusLactobacillus spp.Streptococcus lutetiensisLactococcus garvieaeStreptococcus mitisLeptotrichia spp.Streptococcus mutansMicrococcus spp.Streptococcus parasanguinisMoraxella catarrhalisStreptococcus salivariusMoraxella osloensisViridians Group StreptococcusNeisseria mucosaNeisseria siccaPaenibacillus spp.Propionibacterium spp.Rothia* spp.spp = species.Predefined commensal organismsspp = species.

    Results

    A total of 5,296 paired BC sets were evaluated, with 4,411 classified as STS and 885 as MTS. A total of 343 pairs, including 6.3% (278/4,411) of STS and 7.3% (65/885) of MTS, grew only commensal organisms in at least one of the two BC sets (Table 2). Furthermore, 28.1% (78/278) of those STS pairs and 18.5% (12/65) of those MTS pairs grew commensal organisms in both BC sets (P = .11) (Table 2). These findings were consistent when analyzed separately for ED and non-ED settings (Table 2).Table 2.BC positivity for commensal organisms only, as a function of positive BC with single-time stamp (STS) vs multiple-time stamps (MTS) and patient locationSingle-time stamps (STS); n (% of total)Time stamps Within 1 Hour (MTS); n (% of total)Total; n (% of total)All units1 of 2 BC Positive200 (71.9)53 (81.5)253 (73.8)2 of 2 BC Positive78 (28.1)^12 (18.5)^90 (26.2)Total27865343Emergency department1 of 2 BC Positive158 (71.5)39 (81.3)197 (73.2)2 of 2 BC Positive63 (28.5)^^9 (18.8)^^72 (26.8)Total22148269Non-ED inpatient units1 of 2 BC Positive42 (73.7)14 (82.4)56 (75.7)2 of 2 BC Positive15 (26.3)#3 (17.6)#18 (24.3)Total571774Commensal definition based on microbiology laboratory data; ^ P = .11; ^^ P = .17; # P = .46; BC positive = blood culture positive for a commensal organism.BC positivity for commensal organisms only, as a function of positive BC with single-time stamp (STS) vs multiple-time stamps (MTS) and patient locationCommensal definition based on microbiology laboratory data; ^ P = .11; ^^ P = .17; # P = .46; BC positive = blood culture positive for a commensal organism.

    Discussion

    When two sets of BCs were drawn within one hour of each other, the rate at which both sets were positive for commensal organisms was higher in BCs with STS compared to BCs with MTS, although this did not reach statistical significance.This study expands upon the literature regarding single-site versus multi-site sampling blood culture techniques. Mcleod et al. identified the collection of at least two BCe sets from the same site as a frequent problem in the ED, with a reported 15.13% of samples being drawn in this fashion.6 Yu et al. and Ekwall-Larson et al. reported no statistically significant difference in contamination between multi-site sampling and single-site sampling in Swedish EDs.7,8 Our study demonstrates that multi-site sampling (or BC with MTS) may yield higher rates of commensal organism culture positivity in two of two BC sets.This analysis had limitations. Most notably, we assumed that paired BC sets with identical time stamps represent single-site sampling, which might not have been the case in all instances and would have resulted in potential misclassification bias. We also did not adjudicate whether patients had true bacteremia with commensal organisms. Patients with more comorbidities are likely to have difficult IV access, making single-site sampling BC more likely in this sicker group. Other limitations include a small sample size in a single-center, and the fact that we did not control for which person was drawing the cultures.This study highlights the importance of utilizing optimal blood culture sampling techniques to best inform clinical decision-making and improve patient care. Although underpowered to demonstrate a statistically significant association, the clinical implication of this study is that by increasing the likelihood of recovering commensal organisms in two sets of BCs, single-site sampling practices may increase the risk of misclassifying patients as having true bacteremia, potentially leading to unnecessary antibiotic use and increased length of stay.4 Importantly, this study highlights the need for efficient and accurate documentation methods embedded within the EMR regarding the anatomic sites and timing of BC sampling. One way to do this is to raise awareness amongst nursing staff and phlebotomists regarding the potential adverse impact of inaccurate documentation on patient care. Moreover, collaborating with information technology staff on EMR “nudges” and streamlining documentation can be utilized to improve blood culturing processes.
    AADr. Ahmed Abdul Azim

    Dr. Ahmed Abdul Azim

    Infectious Disease · Antibiotic Stewardship
    Rutgers, The State University of New Jersey
    ahmedabdulazim