ORIGINAL RESEARCH

Short-Course Versus Long-Course Systemic AntibioticTreatment for Uncomplicated Intravascular Catheter-Related Bloodstream Infections due to Gram-NegativeBacteria, Enterococci or Coagulase-NegativeStaphylococci: A Systematic Review

Severin Muff . Alexis Tabah . Yok-Ai Que . Jean-Francois Timsit .

Leonard Mermel . Stephan Harbarth . Niccolo Buetti

Received: March 29, 2021 / Accepted: May 22, 2021 / Published online: June 25, 2021� The Author(s) 2021

ABSTRACT

Introduction: The optimal duration of systemicantimicrobial treatment for catheter-relatedbloodstream infections (CRBSI) is unknown. In

this systematic review, we aimed to assess theefficacy of short-course treatment for CRBSI dueto Gram-negative bacteria, coagulase-negativestaphylococci and enterococci.Methods: We systematically searched the elec-tronic bibliographic databases MEDLINE,EMBASE and Cochrane Library for studies pub-lished before February 2021. All studies thatinvestigated the duration of adequate systemicantibiotic treatment in adult patients withuncomplicated intravascular catheter infectionsdue to Gram-negative bacteria, coagulase-nega-tive staphylococci or enterococci were eligiblefor inclusion. Studies including concomitanttreatment with antibiotic lock therapy wereexcluded. The primary outcomes were clinicalfailure/cure, mortality and microbiologic-con-firmed relapse.Results: Seven retrospective cohort studies andone case-cohort study met the inclusion crite-ria. No randomized controlled studies metinclusion criteria. The quality of the includedstudies was low (n = 7) to moderate (n = 1). Nosignificant differences were observed regardingmortality and microbiological relapse betweenshort-course and long-course systemic antibi-otic treatment in patients with CRBSI due tocoagulase-negative staphylococci or Gram-neg-ative bacteria. No association was foundbetween mortality and treatment duration inthe two studies assessing enterococcal CRBSI.Conclusion: The limited data available suggeststhat shorter systemic antibiotic treatment

Supplementary Information The online versioncontains supplementary material available at https://doi.org/10.1007/s40121-021-00464-0.

S. Muff � S. Harbarth � N. Buetti (&)Infection Control Program and WHO CollaboratingCentre on Patient Safety, University of GenevaHospitals and Faculty of Medicine, Rue Gabrielle-Perret-Gentil 4, 1205 Geneva, Switzerlande-mail: niccolo.buetti@gmail.com

A. TabahICU, Faculty of Medicine, Redcliffe Hospital,University of Queensland, Brisbane, QLD, Australia

Y.-A. QueInselpital, Bern University Hospital, University ofBern, Bern, Switzerland

J.-F. TimsitTeam DeSCID, University of Paris, INSERM IAME,U1137, Paris, France

J.-F. TimsitMedical and Infectious Diseases Intensive Care Unit,AP-HP, Bichat-Claude Bernard University Hospital,46 rue Henri Huchard, 75877 Paris Cedex, France

L. MermelDivision of Infectious Diseases, Department ofMedicine, Rhode Island Hospital, Alpert MedicalSchool of Brown University, 593 Eddy St.,Providence, RI, USA

Infect Dis Ther (2021) 10:1591–1605

https://doi.org/10.1007/s40121-021-00464-0

duration may be sufficient for uncomplicatedCRBSI. Further well-designed prospective stud-ies are needed to confirm these findings.Trial Registration Number: CRD42021224946(PROSPERO)

Keywords: Catheter-related bloodstreaminfection; Coagulase-negative staphylococci;CRBSI; Enterococcal; Enterococcus; Gram-negative; Intravascular catheter infection

Key Summary Points

The optimal duration of antibiotic therapyfor intravascular catheter-relatedbloodstream infections is unknown andcurrent recommendations are mainlybased on expert opinions.

The aim of this systematic review was toassess short-course versus long-coursetreatment for catheter-relatedbloodstream infections due to Gram-negative bacteria, enterococci orcoagulase-negative staphylococci.

In the studies assessed, patients withshort-course treatment had similarmortality rates, clinical cure rates andmicrobiological relapse rates as patientswith long-course treatment.

Prospective studies are needed todetermine the optimal antimicrobialtreatment duration for catheter-relatedbloodstream infections.

DIGITAL FEATURES

This article is published with digital features,including a summary slide, to facilitate under-standing of the article. To view digital featuresfor this article go to https://doi.org/10.6084/m9.figshare.14627559.

INTRODUCTION

Intravascular catheters are used for the admin-istration of fluids, blood products and medica-tion, to monitor haemodynamic parametersand for haemodialysis. They are essential for thecare of hospitalized patients, particularly in thecase of critical illness. Unfortunately, intravas-cular catheters are also possible ports of entryfor pathogens into the bloodstream. In 2019,over 18,000 cases of central line associatedbloodstream infections (CLABSI) were reportedin acute care hospitals in the USA [1]. In Europe,36.5% of intensive care unit (ICU)-acquiredbloodstream infections (BSIs) were linked tointravascular catheters and rates of CLABSIranged from 1.7 to 4.8 episodes per 1000catheter days [2]. Catheter-related bloodstreaminfections (CRBSI) were associated with sub-stantial morbidity, increased hospital length ofstay [3, 4], mortality [4, 5] and an estimatedattributable cost of US $45,814 per event [6].According to recently published Centers forDisease Control and Prevention (CDC) data,Gram-negative bacteria (GNB), enterococci andcoagulase-negative staphylococci (CoNS) wereamong the most frequently detected bacterialpathogens associated with such infections [7].

Research to determine the appropriateantibiotic treatment duration is part of theUnited States National Action Plan for Com-bating Antibiotic-Resistant Bacteria 2020–2025[8]. Recent literature derived from randomizedcontrolled trials (RCTs) suggested that short-course treatment for hospital and community-acquired Gram-negative bloodstream infectionsis non-inferior to longer courses of treatment[9, 10]. Similar results were confirmed by tworecent systematic reviews [11, 12]. Nevertheless,little is known about the optimal duration ofantimicrobial treatment for CRBSI. The Infec-tious Disease Society of America (IDSA) andother current guidelines recommended treat-ment durations of 5–14 days for uncomplicatedenterococcal CRBSI, 7–14 days for Gram-nega-tive CRBSI and 5–7 days for CoNS-CRBSI[13–16]. However, these recommendations weremainly based on expert opinions and, to ourknowledge, no recent literature review has been

1592 Infect Dis Ther (2021) 10:1591–1605

performed. Therefore, we conducted a system-atic review with the objective to evaluate if theduration of systemic antibiotic treatment inpatients with uncomplicated CRBSI due toGNB, enterococci, or CoNS influenced the ratesof mortality, clinical cure/failure or microbio-logical relapse.

METHODS

This study was performed according to thePreferred Reporting Items for Systematic reviewsand Meta-Analyses (PRISMA) statements andwas prospectively registered within the inter-national prospective register of systematicreviews (PROSPERO, ID CRD42021224946) on15 December 2020 [17]. This article is based onpreviously conducted studies and does notcontain any new studies with human partici-pants or animals performed by any of theauthors.

Search Strategy, Study Screeningand Selection

We conducted a systematic search of the elec-tronic bibliographic databases MEDLINE(Pubmed, 1949 to date), EMBASE (Elsevier, 1974to date) and the Cochrane Library (includingthe Cochrane Central Register of ControlledTrials). Additionally, we manually searchedthroughout the European Congress of ClinicalMicrobiology and Infectious Diseases (ECCMID2008–2019) and ID-WEEK (2014–2019) confer-ence websites and clinicaltrials.gov to detectgrey literature and unpublished studies. The fullsearch strategies are available in the supple-mentary material. Briefly, the searches wereconstructed with controlled vocabulary termsand keywords describing ‘‘intravascular catheterinfection’’, ‘‘antibiotic therapy’’, ‘‘duration oftherapy’’, ‘‘Gram-negative bacilli’’, ‘‘enterococ-cus’’ and ‘‘coagulase-negative staphylococci’’.The controlled vocabulary terms were adaptedfor each database. We placed no restrictions onlanguage, publication status or publicationperiod. The final search update was performedon 15 February 2021. The results were importedinto a literature management software to

exclude duplicate publications. Then, title andabstract screening and subsequently full-textscreening were performed independently bytwo authors (Severin Muff and Niccolo Buetti),with any disagreement resolved through dis-cussion between the authors. To manage theselection process, we used a web-based screen-ing and data extraction tool [18]. All study typeswith the exception of case reports, letters andcomments were eligible for inclusion.

We included studies that investigated theduration of adequate antibiotic treatment inadult patients with uncomplicated intravascularcatheter infections due to GNB, enterococci orCoNS. Studies assessing treatment duration forother microorganisms (e.g. Staphylococcus aureusor Candida species) were excluded because theynecessitate longer antibiotic treatment dura-tions (i.e. at least 14 days). Of note, Staphylo-coccus lugdunensis was not included amongCoNS since it can cause an aggressive clinicalcourse, and as a consequence, longer treatmentdurations are recommended [15, 16, 19, 20]. Weexcluded studies investigating lock therapieswith systemic antimicrobials because locktherapy may require more prolonged systemicantibiotic courses, and our target populationwas uncomplicated intravascular catheterinfection. Intravascular catheter infections weredefined as uncomplicated if clinical andmicrobiological signs of infections resolvedwithin 72 h after initiation of antibiotic treat-ment and if complications like metastaticinfection, endocarditis or suppurative throm-bophlebitis were absent. Studies includingantimicrobial-coated catheters were excluded.Studies exclusively including paediatric patientswere excluded. Studies with no age restrictionswere included if the majority of the study pop-ulation was adult.

Data Extraction

For evidence synthesis we used a standardizedform to extract data from the included studies.We extracted data on study design, setting,participants, intervention groups (i.e. treatmentduration), catheter removal and outcome

Infect Dis Ther (2021) 10:1591–1605 1593

measurements. Information about risk of biasassessment was extracted with a separate form.

Intervention and Outcomes

The interventions assessed by this review wereshort-course versus long-course adequateantimicrobial treatment for uncomplicatedintravascular catheter-related infections due toGNB, CoNS or enterococci. Duration of short-and long course treatment was predefined bythe authors of the selected studies. Primaryoutcomes were mortality (in-hospital or day-30mortality), clinical cure/clinical failure andmicrobiologically confirmed relapse. Clinicalcure was defined as resolution of symptomsand/or the significant reduction in inflamma-tory parameters. Clinical failure was defined aspersistence of sign and symptoms of infectionsand/or lack of significant reduction in inflam-matory parameters. We assumed that patientsnot fulfilling the criteria for clinical failure were‘clinically cured’. Microbiological relapse wasdefined as reoccurrence of the same pathogen inat least one blood culture after an initial clinicalresponse to treatment within the follow-upperiod. Secondary outcomes were antibiotic-as-sociated adverse events.

Risk of Bias

To assess the quality (including the risk of bias)of cohort and case–control studies we used theNewcastle–Ottawa scale [21]. The studies wereevaluated by two authors (Severin Muff andNiccolo Buetti) independently; any disagree-ment was resolved through discussion. Scores of0–3, 4–6 and 7–9 were considered as low, mod-erate and high quality, respectively.

Study Synthesis

Since only retrospective observational studieswith high levels of clinical and methodologicalheterogeneity were retrieved, we did not per-form a quantitative synthesis of data and wesummarized our findings in the form of a nar-rative synthesis.

RESULTS

Selection

The review process is summarized in Fig. 1. Oursearch strategy identified 1491 studies, of which236 duplicates were removed. Of the remaining1255 studies, 11 were potentially eligible. Afterfull-text screening, three of 11 studies wereexcluded because of insufficient assessment oftreatment duration (n = 2) and missing assess-ment of intravascular catheter infections(n = 1). Finally, eight studies were included inthe review.

Study Characteristics

The main characteristics of the included studiesare presented in Table 1. Among eight includedstudies, seven were cohort studies [22–28]. Onestudy described a retrospective cohort (i.e.‘‘phase 1’’) and a prospective case-cohort (i.e.‘‘phase 2’’). The cases in the phase 2 were mat-ched with controls of phase 1 and we onlyincluded the results of the case-cohort phase[29]. The total number of included patients was780 with numbers ranging from 44 to 188patients per study. All studies were conductedin tertiary-care hospitals and only hospitalizedpatients were included. All cohort studiesassessed the whole database of their institutionsto identify patients fulfilling the inclusion cri-teria. Adult patients were included in fourstudies [24, 26, 27, 29], whereas four studiesincluded patients of all ages [22, 23, 25, 28].Catheter removal was mandatory for inclusionin five studies [22–24, 27, 29]. No study assessedantibiotic-associated adverse events.

The risk of bias was assessed by the New-castle–Ottawa scale [21] (Table 2). According toour study assessment seven studies [23–29] wereof low methodological quality while one study[22] was of moderate quality.

Gram-Negative Intravascular CatheterInfections

Three studies, including a total of 157 patients,assessed treatment duration for uncomplicated

1594 Infect Dis Ther (2021) 10:1591–1605

intravascular catheter-related infections due toGram-negative bacteria [22, 24, 29]. In all ofthese studies, catheter removal was mandatoryfor inclusion. Two cohort studies compared upto 7 days to more than 7 days of treatment[22, 24]. The differences between the short- andlong-course group for clinical cure, relapse and30-day mortality were not significantly differentin both studies. The case-cohort study com-pared patients receiving at least 14 days withpatients receiving 7 days of antibiotic treatment[29]. No significant differences between thetreatment groups regarding death and relapsewere observed. Clinical cure was notinvestigated.

CoNS Intravascular Catheter Infections

Three cohort studies compared short- versuslong-course treatment for CRBSI due to coagu-lase-negative staphylococci [23, 27, 28]. Treat-ment durations varied from no antibiotictreatment to less than 10 days of treatment inthe short-course groups and from more than3 days to more than 10 days in the long-coursegroups.

The two more recent studies were relativelysimilar: Hebeisen et al. compared no antibiotictreatment with treatment for at least 5 days [27]and San-Juan et al. compared at most 3 days

Fig. 1 Study flow diagram. *Studies did not meet inclusion criteria

Infect Dis Ther (2021) 10:1591–1605 1595

Table

1Studycharacteristics,interventionsandoutcom

es

Reference

Stud

ydesign;stud

yperiod

;coun

try

No.

ofpatients

includ

ed

Setting

Patientscharacteristics

Catheterremoval

Intervention

s

Age

(years),

mean

(–SD

)

Sex(m

ale

gend

erin

%)

Antibioticcourse

bydu

ration

Short-

course

Lon

g-course

Gram-negative

bacteria

Ruiz-

Ruigomez

etal.2

020

[22]

Retrospective

cohort

study;2012–2

018;

Spain

54patients

Tertiary-care

hospital

57.9 (±

15.9)

59.3%

Catheterremoval

mandatory

forinclusion

B7days

[7days

Lee

etal.2

021

[24]

Retrospective

cohort

study;2007–2

016;

SouthKorea

59patients

Twotertiary-

care

hospitals

69(59–

73)a

52.5%

Catheterremoval

mandatory

forinclusion

B7days

[7days

Surapatet

al.

2020

[29]

Case-coho

rtstudy

2016–2

017;

Thailand

44patients

Tertiary-care

hospital

67(18–

81)b

47.8%

Catheterremoval

mandatory

forinclusion

7days

C14

days

Coagulase-negativestaphylococci

Hebeisenet

al.

2019

[27]

Retrospective

cohort

study;2008–2

016;

Switzerland

184patients

Tertiary-care

hospital

61(51–

67)a

70%

Catheterremoval

mandatory

forinclusion

0days

[5days

SanJuan

etal.

2019

[23]

Retrospective

cohort

study;2012–2

017;

Spain

79patients

Tertiary-care

hospital

53(±

24.8)

53.2%

Catheterremoval

mandatory

forinclusion

B3days

[3days

Raadet

al.

2009

[28]

Retrospective

cohort

study;2005–2

007;

USA

188patients

University

cancer

center

56(1–8

7)b

62%

Cathetersof

110patients

wereremoved

orexchanged

\10

days

C10

days

Enterococci

Sand

oeet

al.

2002

[25]

Retrospective

cohort

study;1998–2

000;

UK

61patients

Tertiary-care

hospital

63(0.2–8

0)b

57%

48of

61catheterswere

removed

Median5days

(mean

6.4,

range0–

21)

Marschalletal.

2013

[26]

Retrospective

cohort

study;2006;USA

111patients

Tertiary-care

hospital

58.2

±15.3

50.5%

82of

111patientshad

theircathetersremoved

Meanduration

10±

8days

1596 Infect Dis Ther (2021) 10:1591–1605

Table

1continued

Reference

Outcomeby

duration

,no

.of

patients/total

no.of

patients

Add

itional

comment

All-causemortality

Clin

ical

cure

Relapse

Short-course

Lon

g-course

Short-course

Lon

g-course

Short-course

Lon

g-course

Gram-negativebacteria

Ruiz-Ruigomez

etal.2

020[22]

30-day

mortality

3/23

(13%

)

30-day

mortality

1/29

(3%)

18/23(78%

)23/29(79%

)1/23

(4%)

0/29

(0%)

Assessm

entof

outcom

esfor

52of

54

included

patients

Lee

etal.2

021

[24]

30-day

mortality

4/30

(13%

)

30-day

mortality

6/29

(21%

)

22/30(73%

)18/29(62%

)1/30

(3%)

1/29

(3%)

Surapatet

al.

2020

[29]

1/11

(9%)

4/33

(12%

)Not

performed

Not

performed

0/11

(0%)

1/33

(3%)

Wedescribed

onlythe

phase2of

thestudy

Coagulase-negativestaphylococci

Hebeisenet

al.

2019

[27]

Hospitalmortality

6/32

(19%

)

Hospitalmortality

12/140

(9%)

32/32c

(100%)

111/140c

(79%

)0/32

(0%)

8/140(6%)

Assessm

entof

outcom

esfor

172of

184

included

patients

SanJuan

etal.

2019

[23]

30-day

mortality

1/25

(4%)

30-day

mortality

3/54

(6%)

24/25d

(96%

)36/54d

(67%

)3/25

(12%

)10/54(19%

)22.8%

paediatric

patients

Raadet

al.2

009

[28]

Not

performed

Not

performed

38/41(93%

)137/147(93%

)1/32

(3%)

16/121

(13%

)Included

paediatric

patients

Infect Dis Ther (2021) 10:1591–1605 1597

(which included patients with no activeantibiotic treatment) with more than 3 days oftreatment [23]. Catheter removal was manda-tory in these two studies. The number ofpatients with clinical cure was significantlylower in the long-course treatment groups inboth studies. Hebeisen et al. reported that allpatients with short-course treatment (32/32)were clinically cured in contrast to only 79%(111/140) in the long-course group (P = 0.01).San-Juan et al. reported similar results with 96%(24/25) clinically cured patients in the short-course treatment group and only 67% (36/64)clinically cured patients in the long-coursetreatment group (P = 0.07). However, patientsin the long-course groups were significantlymore immunocompromised [27] or moreseverely ill [23]. No significant differences inrelapse or mortality were found. The authorsconcluded that catheter removal may be suffi-cient to treat CoNS-CRBSI in selected patients.

Raad et al. primarily assessed catheter reten-tion versus removal and included only patientswith cancer [28]. The study revealed similarclinical cure and relapse rates for prolongedtherapy (i.e. at least 10 days) and shorter ther-apy durations (i.e. less than 10 days).

Enterococcal Intravascular CatheterInfections

Only two cohort studies investigated entero-coccal CRBSI [25, 26]. Both primarily investi-gated catheter retention versus catheterremoval and/or catheter management. There-fore, patients with catheter retainment wereincluded in both studies. No studies performedan exhaustive comparison between short versusand longer antimicrobial treatment duration.Treatment duration was evaluated using uni-variate analyses. The first study assessed 61enterococcal CRBSIs. Antimicrobial treatmentduration did not impact cure, treatment failureor relapse [25]. Moreover, an analysis of patientstreated by catheter removal plus adequateantibiotic therapy and patients treated bycatheter removal without (adequate) antibiotictherapy did not show statistically significantdifferences for these outcomes. The second

Table

1continued

Reference

Outcomeby

duration

,no

.of

patients/total

no.of

patients

Add

itional

comment

All-causemortality

Clin

ical

cure

Relapse

Short-course

Lon

g-course

Short-course

Lon

g-course

Short-course

Lon

g-course

Enterococci

Sand

oeet

al.

2002

[25]

Durationof

antimicrobialtherapywas

notassociated

withmortality

Durationof

antimicrobialwas

not

associated

withclinicalcure

Durationof

antimicrobialtherapywas

notassociated

withmicrobiological

relapse

14%

paediatric

patients

Marschallet

al.

2013

[26]

Noassociationbetweenmortality

andtreatm

entduration

foun

d

Not

performed

Not

performed

aMedian(interquartilerange)

bMedian(range)

cStatistically

significant

difference

(pvalue=0.01)

dMarginally

statisticsignificant

difference

(pvalue=0.07).Allotherdifferencesin

outcom

esbetweenshort-andlong-coursegroups

werenotstatistically

significant

SDstandard

deviation,

USA

UnitedStates

ofAmerica,UK

UnitedKingdom

1598 Infect Dis Ther (2021) 10:1591–1605

Table2

Qualityassessment

Coh

ortstud

ies

Ruiz-

Ruigomez

etal.2020

[22]

Lee

etal.

2021

[24]

Hebeisen

etal.

2019

[27]

San-

Juan

etal.

2019

[23]

Raad

etal.

2009

[28]

Sand

oeet

al.

2002

[25]

Marschall

etal.2

013

[26]

Case-coho

rtstud

ies

Surapat

etal.

2020

[29]

Totalscore

42

32

32

23

Selection

Representativenessof

the

exposedcohort

––

––

––

–Selection

Adequacyof

case

defin

ition

–

Selectionof

thenon-

exposedcohort

**

*–

*–

–Representativeness

ofthecases

–

Ascertainmentof

exposure

––

––

––

–Selectionof

controls

–

Dem

onstration

that

outcom

eof

interest

was

notpresentat

startof

study

**

**

**

*Definition

of

controls

*

Com

parability

––

––

––

–Com

parability

*–

Outcomes

Assessm

entof

outcom

e–

––

––

––

Exposure

Ascertainmentof

exposure

–

Was

follow-uplong

enough

foroutcom

es

tooccur?

*–

**

**

*Samemethodof

ascertainm

ent

forcasesand

controls

*

Adequacyof

followup

of

cohorts

*–

––

––

–Non-responserate

–

*The

studyhasmet

thecriteriaforthisdomainof

theNew

castle–O

ttaw

ascale

–The

studyhasnotmet

thecriteriaforthisdomainof

theNew

castle–O

ttaw

ascale

Infect Dis Ther (2021) 10:1591–1605 1599

study included 111 enterococcal CRBSIs and didnot find an association between the treatmentduration and mortality [26].

DISCUSSION

Over the last 2 decades, a substantial reductionin CLABSI rates has been achieved owing to thedevelopment and implementation of effectiveprevention measures [30]. Our systematicreview identified only eight observationalstudies which assessed treatment duration forGram-negative, CoNS and enterococcal CRBSI.Interestingly, no RCT that compared short-course versus long-course antibiotic treatmentdurations for intravascular catheter infectionswas retrieved. High-quality research on themanagement of intravascular catheter infec-tions is scarce and guidelines rely on expertopinion. The few studies addressing the topicincluded a very limited number of patients and,probably because of the retrospective nature ofthese studies, short- and long-course groupswere not well balanced. Notwithstanding, theydid not identify significant differences in mor-tality or relapse rates between short- and long-course treatment groups.

Optimizing use of antibiotics is an essentialstep in fighting against antibiotic resistance [8].Our results revealed that unadjusted outcomesfor short-course treatment (i.e. 7 days) andlonger courses of treatment were similar amonguncomplicated Gram-negative CRBSI[22, 24, 29]. Our findings were supported by tworecent RCTs on treatment of uncomplicatedGram-negative bacteraemia including catheter-related urinary tract infection, which demon-strated similar outcomes for short- and long-course antibiotic therapy [9, 10]. In light ofthese considerations, a treatment duration of7 days after source control (i.e. catheterremoval) is a reasonable recommendation forGram-negative CRBSI.

CoNS, traditionally considered typical skincommensals, have emerged as common causesof nosocomial infections [31]. Internationalguidelines recommended a treatment durationof 5–7 days for CoNS-CRBSI [14–16]. The treat-ment durations were shorter than those

recommended (i.e. 0 days and less than 3 days)in two of the studies we included in this review[14–16, 23, 27]. Relapse and mortality rates inthe study groups without treatment or withshort treatment duration were low. Curiously,the clinical cure rate was significantly lower inthe long-course groups [23, 27]; however,patients in the long-course groups were morepredisposed to an unfavourable course (i.e.immunosuppressed, haematological cancer,higher Pitt bacteraemia score). Overall, a veryshort-course (i.e. at most 3 days) of antibiotictreatment should be therefore considered forCoNS-CRBSI. The risk of unfavourable clinicalcourses in patients with severe underliningconditions is most likely considerably higherand they therefore may need longer antibiotictherapy durations.

Little is known about the optimal durationof treatment for enterococcal intravascularcatheter infections. The lack of associationbetween treatment duration and mortality, aswell as clinical cure and microbiological relapsein one study [25], implied that longer treatmentdid not improve outcomes and short treatmentduration may be considered for some patientswith this infection. However, RCTs assessingthe optimal treatment duration are needed todevelop well-founded recommendations.

Current guidelines and expert statementsrecommended treatment durations of 7–14 daysfor Gram-negative CRBSI, 5–7 days for CoNSand 5–14 days for enterococcal CRBSI[13, 14, 16, 19, 20, 32] (Table 3). As the treat-ment duration depends on several factors (e.g.host, causative pathogen, complications andthe type of catheter), the decision on how longto treat a patient should be individualized.Further research to determine the optimum foreach patient is urgently needed. In the interim,we discourage prolonged therapy for uncom-plicated Gram-negative CRBSI (i.e. longer than7 days) and we recommend considering veryshort treatment for uncomplicated CoNS-CRBSIafter catheter removal. Longer antibiotic treat-ment durations may be needed for more com-promised patients. Optimal duration oftreatment for uncomplicated enterococcalCRBSI remains an unresolved issue and a treat-ment of 7–14 days appears to be reasonable [33].

1600 Infect Dis Ther (2021) 10:1591–1605

Table3

Summaryof

recommendation

sandexpertopinions

First

author

Journal

Year

Setting

Recom

mendedtreatm

entdu

ration

Whenprolon

gedtherapy?

Gram-negative

CRBSI/C

LABSI

Coagulase-

negative

staphylococcal

CRBSI/

CLABSI

Enterococcal

CRBSI/

CLABSI

German

guidelines

Boll

Ann Hem

atol

2021

Oncology

Pseudomonas

and

Stenotrophom

onas:

C2weeks

5–7days

after

defervescence

5–7days

after

defervescence

Com

plications

(end

ocarditis,

osteom

yelitis)

French

recommendation

s

Tim

sit

Ann In

tensive

Care

2020

ICU

Enterobacteriaceae,

Pseudomonas

aeruginosa,

Acinetobacter

baum

annii:7days

7days

7days

Rem

otecomplications

Expertstatem

ent

Buetti

Semin

Respir

CritC

are

Med

2019

ICU

Enterobacteriaceae:

(5–)

7days

Pseudomonas

aeruginosa,

Acinetobacter

baum

annii:7days

(5–)

7days

(5–)

7days

Persistent

CRBSI,com

plicated

courses(i.e.another

vascular

line

infection,

metastaticabscess,

septicthrombophlebitisor

endocarditis)

Spain

recommendation

s

Chaves

Med Intensiva

2018

ICU

C7days

5–7days

7–14

days

ForCoN

S:10–1

4daysforpatients

withintravasculardevices,

biom

edicaldevicesor

persistent

markersof

inflammationafter

catheter

removal

Internationalexpert

consensus

statem

ent

Tim

sit

Intensive

Care

Med

2018

ICU

7–14

days

5–7days

7–14

days

Persistent

bacteraemia,

complications

relatedto

bacteraemia(i.e.suppurative

thrombophlebitis,endocarditis,

osteomyelitis,m

etastatic

infection)

Infect Dis Ther (2021) 10:1591–1605 1601

Table3

continued

First

author

Journal

Year

Setting

Recom

mendedtreatm

entdu

ration

Whenprolon

gedtherapy?

Gram-negative

CRBSI/C

LABSI

Coagulase-

negative

staphylococcal

CRBSI/

CLABSI

Enterococcal

CRBSI/

CLABSI

Expertstatem

ent

Rupp

InfectDis

Clin

North

Am

2018

All catheters

Pseudomonas

or

MDRGNB:

10–1

4days

Other

GNB:

7–14

days

5–7days

7–14

days

Com

plicated

CRBSI

(i.e.

suppurativethrombophlebitis,

persistent

bacteraemia,

osteom

yelitis,infective

endocarditis)

IDSA

guidelines

(USA

)

Mermel

Clin

Infect

Dis

2009

All catheters

7–14

days

5–7days

or

undercertain

circum

stances

observation

without

antibiotics

7–14

days

Com

plicated

CRBSI

(i.e.

suppurativethrombophlebitis,

osteom

yelitis,infective

endocarditis)

ICUintensivecare

unit,C

RBSI

catheter-related

bloodstream

infection,

CLABSI

centrallineassociated

bloodstream

infection,

USA

UnitedStates,IDSA

Infectious

DiseasesSocietyof

America,GNBGram-negativebacteria,M

DRmultidrug-resistant.

1602 Infect Dis Ther (2021) 10:1591–1605

Our systematic review has several limita-tions. We did not perform a quantitative syn-thesis because of the heterogeneity and lowquality of the studies. We modified our designcompared to the published protocol by includ-ing studies which did not exclude paediatricpatients but studies exclusively including pae-diatric patients were excluded. Paediatric rep-resented a limited part of the included patientsand recommendations for this specific popula-tion cannot be provided. Moreover, our out-come definition of clinical failure/cure wasmodified. Studies of enterococcal CRBSI did notperform a comparison between short- and long-course treatment. As a consequence, an exactevaluation of treatment duration was not pos-sible. We did not include studies using antimi-crobial lock therapy which is recommended inguidelines and in use clinically at some insti-tutions. We did not differentiate studiesinvolving different central venous catheters(e.g. short-term catheters versus long-termcatheters such as tunnelled or implantedcatheter or those used exclusively for dialysis).We did not differentiate Gram-negative infec-tions on the basis of the pathogen involved (e.g.outcomes of CRBSI due to Pseudomonas spp.versus Klebsiella spp). Finally, we included onlyuncomplicated CRBSI and therefore no conclu-sion can be drawn on complicated intravascularcatheter infections.

CONCLUSION

Our review suggests that shorter antibioticduration for uncomplicated intravascularcatheter infections due to CoNS and Gram-negative bacilli can be implemented when theinfected catheter was removed. However, well-designed studies are urgently needed to confirmthis recommendation.

ACKNOWLEDGEMENTS

Funding. Niccolo Buetti is currently receiv-ing a Mobility grant from the Swiss NationalScience Foundation (Grant No.

P4P4PM_194449). The Journal’s Rapid ServiceFee was funded by Swiss National ScienceFoundation.

Authorship. All named authors meet theInternational Committee of Medical JournalEditors (ICMJE) criteria for authorship for thisarticle, take responsibility for the integrity ofthe work as a whole, and have given theirapproval for this version to be published.

Authors’ Contributions. Niccolo Buetti andSeverin Muff contributed to the study concep-tion and design. Niccolo Buetti and SeverinMuff were responsible for the literature searchand the data collection. All authors analysedand interpreted the data. Niccolo Buetti andSeverin Muff were the major contributors inwriting the manuscript. Alexis Tabah, Yok-AiQue, Jean-Francois Timsit, Leonard Mermel andStephan Harbarth critically revised the manu-script. All authors read and approved the finalmanuscript.

Disclosures. Severin Muff, Alexis Tabah,Yok-Ai Que, Stephan Harbarth, Niccolo Buettihave disclosed that they do not have conflict ofinterest. Jean-Francois Timsit received fees forlectures from 3M, MSD, Pfizer, and Biomerieux.Jean-Francois Timsit received research grantsfrom Astellas, 3M, MSD, and Pfizer. Jean-Fran-cois Timsit participated in advisory boards of3M, MSD, Bayer Pharma, Nabriva, and Pfizer.Leonard Mermel has served as a consultant forCitius Pharmaceuticals, Marvao Medical, Leo-nard-Meron Biosciences, Destiny Pharma, andNobio unrelated to the submitted work.

Compliance with Ethics Guidelines. Thisarticle is based on previously conducted studiesand does not contain any new studies withhuman participants or animals performed byany of the authors.

Data Availability. The datasets generatedduring and/or analysed during the currentstudy are available from the correspondingauthor on reasonable request. Preliminary datawere submitted (abstract) to the ICPIC confer-ence (Geneva, September 2021).

Infect Dis Ther (2021) 10:1591–1605 1603

Open Access. This article is licensed under aCreative Commons Attribution-NonCommer-cial 4.0 International License, which permitsany non-commercial use, sharing, adaptation,distribution and reproduction in any mediumor format, as long as you give appropriate creditto the original author(s) and the source, providea link to the Creative Commons licence, andindicate if changes were made. The images orother third party material in this article areincluded in the article’s Creative Commonslicence, unless indicated otherwise in a creditline to the material. If material is not includedin the article’s Creative Commons licence andyour intended use is not permitted by statutoryregulation or exceeds the permitted use, youwill need to obtain permission directly from thecopyright holder. To view a copy of this licence,visit http://creativecommons.org/licenses/by-nc/4.0/.

REFERENCES

1. Centers for Disease Control and Prevention. Centralline-associated bloodstream infections: A.R. &Patient Safety Portal. 2019. https://arpsp.cdc.gov/profile/infections/CLABSI. Accessed 11 Jan 2021.

2. European Centre for Disease Prevention and Con-trol. Healthcare-associated infections in intensivecare units - Annual Epidemiological Report for2017. ECDC; 2019 Oct. https://www.ecdc.europa.eu/sites/default/files/documents/AER_for_2017-HAI.pdf.

3. Dimick JB, Pelz RK, Consunji R, Swoboda SM,Hendrix CW, Lipsett PA. Increased resource useassociated with catheter-related bloodstream infec-tion in the surgical intensive care unit. Arch Surg.2001;136:229.

4. Olaechea PM, Palomar M, Alvarez-Lerma F, et al.Morbidity and mortality associated with primaryand catheter-related bloodstream infections incritically ill patients. Rev Esp Quimioter. 2013;26:21–9.

5. Ziegler MJ, Pellegrini DC, Safdar N.Attributable mortality of central line associatedbloodstream infection: systematic review and meta-analysis. Infection. 2015;43:29–36.

6. Zimlichman E, Henderson D, Tamir O, et al. Healthcare-associated infections: a meta-analysis of costs

and financial impact on the US health care system.JAMA Intern Med. 2013;173:2039–46.

7. Novosad SA, Fike L, Dudeck MA, et al. Pathogenscausing central-line-associated bloodstream infec-tions in acute-care hospitals-United States,2011–2017. Infect Control Hosp Epidemiol.2020;41:313–9.

8. The Task Force for Combating Antibiotic-ResistantBacteria. National Action Plan for CombatingAntibiotic-Resistant Bacteria, 2020–2025. 2020.https://aspe.hhs.gov/pdf-report/carb-plan-2020-2025. Accessed 7 Jan 2021.

9. von Dach E, Albrich WC, Brunel A-S, et al. Effect ofC-reactive protein-guided antibiotic treatmentduration, 7-day treatment, or 14-day treatment on30-day clinical failure rate in patients withuncomplicated Gram-negative bacteremia: a ran-domized clinical trial. JAMA. 2020;323:2160.

10. Yahav D, Franceschini E, Koppel F, et al. Sevenversus 14 days of antibiotic therapy for uncompli-cated Gram-negative bacteremia: a noninferiorityrandomized controlled trial. Clin Infect Dis.2019;69:1091–8.

11. Tansarli GS, Andreatos N, Pliakos EE, Mylonakis E.A systematic review and meta-analysis of antibiotictreatment duration for bacteremia due to Enter-obacteriaceae. Antimicrob Agents Chemother.2019;63:e02495-e2518.

12. Li X, Liu C, Mao Z, Li Q, Qi S, Zhou F. Short-courseversus long-course antibiotic treatment in patientswith uncomplicated gram-negative bacteremia: asystematic review and meta-analysis. J Clin PharmTher. 2021;46:173–80.

13. Boll B, Schalk E, Buchheidt D, et al. Central venouscatheter-related infections in hematology andoncology: 2020 updated guidelines on diagnosis,management, and prevention by the InfectiousDiseases Working Party (AGIHO) of the GermanSociety of Hematology and Medical Oncology(DGHO). Ann Hematol. 2021;100:239–59.

14. Timsit J-F, Baleine J, Bernard L, et al. Expert con-sensus-based clinical practice guidelines manage-ment of intravascular catheters in the intensive careunit. Ann Intensive Care. 2020;10:118.

15. Mermel LA, Allon M, Bouza E, et al. Clinical prac-tice guidelines for the diagnosis and managementof intravascular catheter-related infection: 2009update by the Infectious Diseases Society of Amer-ica. Clin Infect Dis. 2009;49:1–45.

16. Chaves F, Garnacho-Montero J, Del Pozo JL, et al.Diagnosis and treatment of catheter-related blood-stream infection: clinical guidelines of the Spanish

1604 Infect Dis Ther (2021) 10:1591–1605

Society of Infectious Diseases and Clinical Micro-biology and (SEIMC) and the Spanish Society ofSpanish Society of Intensive and Critical CareMedicine and Coronary Units (SEMICYUC). MedIntens. 2018;42:5–36.

17. Moher D, Liberati A, Tetzlaff J, Altman DG, PRISMAGroup. Preferred reporting items for systematicreviews and meta-analyses: the PRISMA statement.PLoS Med. 2009;6:e1000097.

18. Ouzzani M, Hammady H, Fedorowicz Z, Elma-garmid A. Rayyan—a web and mobile app for sys-tematic reviews. Syst Rev. 2016;5:210.

19. Rupp ME, Karnatak R. Intravascular catheter-relatedbloodstream infections. Infect Dis Clin N Am.2018;32:765–87.

20. Buetti N, Timsit J-F. Management and prevention ofcentral venous catheter-related infections in theICU. Semin Respir Crit Care Med. 2019;40:508–23.

21. Wells G, Shea B, O’Connell D, et al. The New-castle–Ottawa Scale (NOS) for assessing the qualityof non-randomized studies in meta-analysis. 2000.http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp. Accessed 19 Jan 2021.

22. Ruiz-Ruigomez M, Fernandez-Ruiz M, San-Juan R,et al. Impact of duration of antibiotic therapy incentral venous catheter-related bloodstream infec-tion due to Gram-negative bacilli. J AntimicrobChemother. 2020;75:3049–55.

23. San-Juan R, Martınez-Redondo I, Fernandez-Ruiz M,et al. A short course of antibiotic treatment is safeafter catheter withdrawal in catheter-relatedbloodstream infections due to coagulase-negativestaphylococci. Eur J Clin Microbiol Infect Dis.2019;38:977–83.

24. Lee Y-M, Moon C, Park K-H. Is short-term antibiotictherapy safe in patients with central venous cathe-ter-related Gram-negative bacteraemia? J Antimi-crob Chemother. 2021;76:539–41.

25. Sandoe JAT, Witherden IR, Au-Yeung H-KC, Kite P,Kerr KG, Wilcox MH. Enterococcal intravascularcatheter-related bloodstream infection:

management and outcome of 61 consecutive cases.J Antimicrob Chemother. 2002;50:577–82.

26. Marschall J, Piccirillo ML, Fraser VJ, Doherty JA,Warren DK. Catheter removal versus retention inthe management of catheter-associated enterococ-cal bloodstream infections. Can J Infect Dis MedMicrobiol. 2013;24:e83–7.

27. Hebeisen UP, Atkinson A, Marschall J, Buetti N.Catheter-related bloodstream infections with coag-ulase-negative staphylococci: are antibiotics neces-sary if the catheter is removed? Antimicrob ResistInfect Control. 2019;8:21.

28. Raad I, Kassar R, Ghannam D, Chaftari AM,Hachem R, Jiang Y. Management of the catheter indocumented catheter-related coagulase-negativestaphylococcal bacteremia: remove or retain? ClinInfect Dis. 2009;49:1187–94.

29. Surapat B, Montakantikul P, Malathum K, Kiert-iburanakul S, Santanirand P, Chindavijak B.Microbial epidemiology and risk factors for relapsein Gram-negative bacteria catheter-related blood-stream infection with a pilot prospective study inpatients with catheter removal receiving short-du-ration of antibiotic therapy. BMC Infect Dis.2020;20:604.

30. Centers for Disease Control and Prevention.National and State Associated Infections ProgressReport. Centers for Disease Control and Prevention.2016. https://www.cdc.gov/HAI/pdfs/progress-report/hai-progress-report.pdf.

31. Heilmann C, Ziebuhr W, Becker K. Are coagulase-negative staphylococci virulent? Clin MicrobiolInfect. 2019;25:1071–80.

32. Timsit J-F, Rupp M, Bouza E, et al. A state of the artreview on optimal practices to prevent, recognize,and manage complications associated withintravascular devices in the critically ill. IntensiveCare Med. 2018;44:742–59.

33. Muff S, Hebeisen U, Timsit J-F, Mermel L, HarbarthS, Buetti N. Treatment duration of enterococcalintravascular catheter-related infections. ClinMicrobiol Infect. 2020;S1198-743X:30717-5.

Infect Dis Ther (2021) 10:1591–1605 1605