Accepted Manuscript

Title: Systematic review of antimicrobial lock therapy for prevention of

central-line-associated bloodstream infections in adult and pediatric cancer
patients

Author: LeAnn B. Norris, Farah Kablaoui, Maggie K. Brilhart, P. Brandon

Bookstaver

PII: S0924-8579(17)30245-5
DOI: [Link] 10.1016/[Link].2017.06.013
Reference: ANTAGE 5185

To appear in: International Journal of Antimicrobial Agents

Received date: 8-7-2016

Please cite this article as: LeAnn B. Norris, Farah Kablaoui, Maggie K. Brilhart, P. Brandon
Bookstaver, Systematic review of antimicrobial lock therapy for prevention of central-line-
associated bloodstream infections in adult and pediatric cancer patients, International Journal of
Antimicrobial Agents (2017), [Link] 10.1016/[Link].2017.06.013.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service
to our customers we are providing this early version of the manuscript. The manuscript will
undergo copyediting, typesetting, and review of the resulting proof before it is published in its
final form. Please note that during the production process errors may be discovered which could
affect the content, and all legal disclaimers that apply to the journal pertain.
Systematic review of antimicrobial lock therapy for prevention of

central-line-associated bloodstream infections in adult and pediatric

cancer patients

LeAnn B. Norris a,*, Farah Kablaoui b, Maggie K. Brilhart c, P. Brandon Bookstaver a

a
College of Pharmacy, University of South Carolina, Columbia, SC, USA
b
Cleveland Clinic, Abu Dhabi, United Arab Emirates
c
Wake Forest Baptist Medical Center, Winston-Salem, NC, USA

*
Corresponding author. Address: Department of Clinical Pharmacy and Outcomes Sciences, College of Pharmacy,

University of South Carolina, 715 Sumter Street, Columbia, SC 29208, USA. Tel.: +1 803 777 7888.

E-mail address: norris@[Link] (L.B. Norris).

ARTICLE INFO

Article history:

Received 8 July 2016

Accepted 15 March 2017

Keywords:

Antimicrobial lock therapy

Central venous catheters

Central-line-associated bloodstream infection

Page 1 of 38
 Catheter-related bloodstream infections

1 Highlights

2  Cancer patients require prolonged venous access, leading to bacteremia.

3  Antibiotic lock therapy (ALT) has been shown to be effective in reducing infections.
4  National guidelines do not currently support the prophylactic use of antibiotic lock therapy in
5 cancer patients.
6  Use of ALT decreased the incidence of central-line-associated blood stream infection
7 (CLABSI) in the majority of studies in cancer patients.
8  Significant differences in definitions of CVC-related infection, dwell times and lock solutions
9 exist in the studies reviewed.
10  Lock therapy may be an adjunct in high-risk cancer patients for the prevention of CLABSI.
11  Higher-quality evidence is needed for specific ALT recommendations.

ABSTRACT

Background: Central venous catheter (CVC) use is commonplace in cancer patients.

Antimicrobial lock therapy (ALT), the instillation of a concentrated antimicrobial solution into

the catheter lumen, is one method for preventing infection among CVCs. This systematic review

discusses the effectiveness and safety of prophylactic ALT in cancer patients with CVCs.

Methods: A literature search was performed using the Medline database and Google Scholar

from inception until April 2016. The following terms were used: ‘antimicrobial lock solution’,

‘antibiotic lock solution’, ‘oncology’, ‘hematology’, ‘pediatrics’, ‘prevention’, ‘cancer’, ‘catheter

related bloodstream infections’, ‘central-line associated bloodstream infection’ (CLABSI) and

‘central venous catheter’. Studies evaluating prophylactic ALT in cancer patients alone were

eligible for inclusion. Case reports, case series and in-vitro studies were excluded.

Results: In total, 78 articles were identified. Following all exclusions, 13 articles (three adult and

10 pediatric) were selected for evaluation. The most common agents utilized were vancomycin

with heparin; ethanol; taurolidine; and minocycline with EDTA. Quality of evidence was

moderate to high in adult studies and low to moderate in pediatric studies. Use of ALT decreased

Page 2 of 38
 the incidence of CLABSI in the majority of studies; however, there were significant differences

in definitions of CVC-related infection, dwell times and lock solutions.

Conclusion: Lock therapy may be an adjunct in high-risk cancer patients for the prevention of

CLABSI; higher quality evidence is needed for specific ALT recommendations.

1. Introduction

In cancer patients, central venous catheters (CVCs) are utilized indefinitely for administration of

chemotherapy, blood products and total parenteral nutrition. Current guidelines from the Centers

for Disease Control (CDC) for the prevention of central-line-associated bloodstream infection

(CLABSI) include handwashing, aseptic technique, site cleansing and impregnated catheter cuffs

[1]. Despite these methods, bloodstream infections (BSIs) secondary to CVCs still occur.

Bacterial or fungal colonization can originate at the catheter hub, followed by biofilm formation

in the lumen disseminating to the bloodstream. In the general population, the National

Healthcare Safety Network reports a mortality rate as high as 12–25% in patients diagnosed with

CLABSI, with approximately 41 000 cases occurring annually [2]. In high-risk populations such

as cancer patients, bacteremia rates range from 11% to 38%, and mortality rates of 9.6% have

been reported in pediatric oncology patients [3–5]. CLABSIs are associated with excess cost and

hospital length of stay, and can delay administration of chemotherapy [6,7]. It is estimated that

10–20% of hematological cancer patients’ CVCs will become infected [8]. CDC considers

hospital-acquired CLABSI to be a preventable infection, classified as a ‘never event’ [9]. This

designation limits the reimbursement of such hospital-acquired infections from agencies

including the Centers for Medicare and Medicaid Services. The annual burden of cost secondary

to CLABSI approximates $2.68 billion.[10].

Page 3 of 38
Skin commensals are often the cause of CVC infections, with 70% of cases involving Gram-

positive organisms. Coagulase-negative staphylococci (CoNS) and Staphylococcus aureus are

the most common pathogens. Gram-negative organisms are identified in 15% of infections, with

fungi and anaerobes each reported to be the cause in approximately 7% of cases [11]. In cancer

patients with febrile neutropenia, Gram-negative infections occur more commonly than in the

general population [12].

1.1. Antimicrobial lock therapy

Biofilms, bacteria colonized within the lumen of the catheter, are able to resist antimicrobial

therapy at standard concentrations. A method commonly known as antimicrobial lock therapy

(ALT) may be an option for both prophylaxis and treatment of CLABSI. ALT involves the

instillation of a highly concentrated antimicrobial solution, often in combination with an

anticoagulant, into the catheter lumen to dwell while the CVC is not in use [13]. In a

prophylactic modality, antimicrobial lock solutions should maintain effective concentrations to

prevent bacterial or fungal adherence, have prolonged stability within the lumen, are compatible

with desired additives and have limited systemic absorption [14]. Several antimicrobial agents

and antiseptic agents have been studied as prophylactic antimicrobial lock solutions.

ALT has been shown to prevent CLABSI, reduce mortality, reduce the burden of healthcare

costs, and decrease hospital length of stay in patients with chronic diseases. In a meta-analysis,

ALT reduced catheter-related bloodstream infections (CRBSIs) and prolonged catheter survival

in hemodialysis-dependent patients [15]. The authors recommended consideration of ALT in

Page 4 of 38
routine clinical practice in conjunction with other prevention modalities [15]. Guidelines of the

Infectious Diseases Society of America (IDSA) for management of CLABSI do not recommend

the routine use of prophylactic ALT, but suggest its use in special circumstances such as patients

with a long-term cuffed or tunneled catheter or port that have a history of multiple CLABSIs

despite optimal maximal adherence to aseptic technique [16]. The National Comprehensive

Cancer Network does not endorse the use of lock solutions in oncology patients due to the

potential risk of resistance following widespread use [17]. According to the American Society of

Clinical Oncology, data are not sufficient to recommend for or against routine use of ALT, and

more research is warranted [18]. To date, the US Food and Drug Administration have not

approved any agents or formulations for the use of ALT for CLABSI prevention or treatment.

Cancer patients are at high risk for CLABSI given the need for treatment requiring sustained

central venous access. Prophylactic ALT has been shown to reduce CLABSI rates, improve

catheter survival, and thus reduce the need for hospitalization. This paper provides a systematic

review of the evidence available for the use of prophylactic ALT in pediatric and adult cancer

patients. Additional insight into compatibility, stability, optimal dwell times and clinical practice

concerns related to ALT are also addressed.

2. Methods

2.1. Search strategy and selection criteria

The Preferred Reporting Items for Systematic Reviews and Meta-Analyses for Protocols was

used to facilitate the preparation and reporting of this systematic review [19]. A literature search

was performed using the Medline database and all open access journals using Google Scholar

Page 5 of 38
from the date of each database’s inception until April 2016. The terms ‘antibiotic lock solution’,

‘antimicrobial lock solution’, ‘oncology’, ‘hematology’, ‘prevention’, ‘cancer’, ‘ethanol’,

‘taurolidine’, ‘central-line associated bloodstream infections’, ‘catheter related bloodstream

infection’ and ‘central venous catheter’ were used. Appropriateness for inclusion in the review

was determined by the following criteria: peer-reviewed study that must include

hematology/oncology patients; outcomes of interest of either infection rate and/or CVC-free

survival rate; and antimicrobial concentrations used in solution plus any additives. Additional

process details including dwell time and administration were included when available.

Observational cohorts, case–control studies and other open label studies were evaluated for

inclusion.

2.2. Study selection

All articles were reviewed by each of the authors to determine inclusion. In-vitro data, animal

models, case reports and studies missing outcomes of interest were excluded. Articles

investigating ALT in a treatment modality were also excluded. Review articles were evaluated

for cross-referencing.

2.3. Data extraction and outcomes

Two authors extracted relevant data (FK, MB) from included studies on patient population,

number of study participants, control and study lock solutions with concentrations, definitions of

CVC-associated infections, reported CLABSI rates per patient, and reported rate of infections

per 1000 CVC-days. If rates were not provided but could be calculated from available data, these

Page 6 of 38
results were included. Two authors independently assessed study quality using the GRADE

(Grading of Recommendations Assessment, Development and Evaluation) criteria [20].

2.4. Search results

The literature search revealed 78 possible articles; 34 articles had defined cancer populations but

following exclusions, 15 articles were deemed appropriate for inclusion [8,21–32]. Two

additional studies were excluded, leaving 13 studies for the final analysis (Fig. 1). One study was

excluded as it was based on a smaller subset of a larger study that was included [33]. Another

study was excluded due to lack of interpretable data of overall infection rates during the study

period [34]. Three studies included adults only [8,21,22] and 10 studies included pediatrics only

(>90% of study population) [23–32]. Among the adult-only studies, two studies were considered

to be high quality [21,22] and one study was considered to be moderate quality [8]. Among the

pediatric studies, two studies were considered to be high quality [23,32], three were considered

to be moderate quality [24,29,31], and five were considered to be low quality [25–28, 30]. Data

were collated and analysed to provide the following summary details and recommendations. The

definitions and terminology used to define BSIs secondary to CVCs (e.g. CRBSI, CLABSI)

varied between studies, but are reported as reflected in each individual study.

3. Prophylactic studies in adult cancer patients

In adults, CVCs are common and provide convenient delivery of long-term chemotherapy.

Recent data have shown infection rates of 2.66 per 1000 line-days in adult hematology/oncology

and bone marrow transplant units, and this ranges from 1.3 to 1.9 per 1000 line-days in

hematology/oncology units alone [35,36]. This review includes one study that evaluated

Page 7 of 38
vancomycin plus heparin (VH) in solution, and two studies that evaluated ethanol, all compared

with a standard control solution of heparin or normal saline (Table 1). Case reports of successful

use of minocycline plus ethylenediaminetetraacetate (mEDTA) as secondary prophylaxis of

CRBSI among adult cancer patients have been reported, but are not included in this review [37].

3.1. Vancomycin-based lock solution

Vancomycin is an antimicrobial with activity against Gram-positive organisms including

methicillin-resistant S. aureus [24]. Carratala et al. performed a randomized, double-blind study

of VH ALT for prevention of Gram-positive CVC-related infections in neutropenic cancer

patients [21]. In total, 117 patients, primarily diagnosed with leukemia (88%), with non-

tunneled, multi-lumen CVCs and predicted to have severe neutropenia (<500 neutrophils) were

recruited. They were randomized to either heparin 10 units/mL (n=57) or vancomycin 25 µg/mL

plus heparin 10 units/mL (n=60) on the day after completion of chemotherapy once neutropenic.

The solution (2.5 mL) was instilled in each lumen and allowed to dwell for 1 h every 2 days, and

was then aspirated and discarded. Cultures of the catheter hub were completed prior to

randomization and then twice weekly. If a febrile episode occurred, blood cultures were drawn,

and ceftazidime and amikacin were given empirically for antibiotic treatment. Significant

colonization of the catheter hub was defined as ≥15 colony-forming units/mL. CRBSI, or

attributable bacteremia, was defined as isolation of identical organisms from the catheter hub and

separate subcutaneous blood draw. The organisms found in the colonized hubs included

Staphylococcus epidermidis (n=7), Staphylococcus capitis (n=1) and Corynebacterium sp. (n=1)

[21]. Significant colonization of the catheter hub occurred in nine of 57 (15.8%) patients treated

with heparin, whereas there was no evidence of colonization in the 60 patients treated with VH

Page 8 of 38
(P=0.001). For the heparin-alone group, four of 57 patients (7%) were reported to have CRBSI

compared with no patients in the VH group (P=0.05). Both groups had similar results for

bacteremia that was not attributable to catheter infection (P=0.35). Catheter occlusion occurred

in six (10.5%) patients receiving heparin vs. four (6.7%) patients receiving VH (P=0.52) [21].

There were no additional adverse events reported and no reports of incompatibility of the VH

solution. Vancomycin with heparin solution has been shown to maintain physical compatibility

and stability for at least 72 h at 27˚C in a separate investigation [38].

3.2. Ethanol-based lock solution

Ethanol is a bactericidal and fungicidal antiseptic, and concerns about development of resistance

or promotion of cross-resistance to other antimicrobial classes are generally lacking [39]. Two

studies were identified that met the criteria and utilized a 70% ethanol-based lock solution [8,22].

Sanders et al. conducted a randomized, double-blind, controlled trial that compared ethanol lock

solution with heparinized saline for the prevention of CLABSI in adult cancer patients [8].

Patients who received chemotherapy and were expected to become neutropenic (<500

neutrophils) were eligible for study inclusion. Patients were randomized to receive lock solutions

with either 70% ethanol (34 prophylactic treatment periods) or heparinized saline (30

prophylactic treatment periods). Each solution (3 mL) was instilled daily in each catheter lumen

to dwell for 2 h before being aspirated. Patients with neutropenic fever were cultured and given

empiric antibiotics (gentamicin and either tazocin or cefepime) until cultures were finalized.

CRBSIs were defined as any BSI occurring where the CVC had been in use during the preceding

48 h. Use of ethanol lock therapy reduced the CVC attributable infection rate significantly

compared with heparin, [0.06/1000 CVC-days vs. 0.31/1000 CVC-days, respectively; P=0.008).

Page 9 of 38
The overall infection rates were 8.8% (3/34) in the ethanol group and 36.7% (11/30) in the

heparin group. Escherichia coli and S. epidermidis were isolated with the greatest frequency.

There were several febrile infections that were not attributable to catheters in both groups [8]. No

adverse events or CVC occlusions were documented by the authors.

Slobbe et al. conducted a second randomized controlled trial comparing 70% ethanol solution

with 0.9% normal saline in adult patients with CVCs for hematological disease [22].

Randomization was catheter based, which allowed for patients to be randomized more than once

if insertion of a new CVC was needed. In the inpatient setting, every lumen of the CVC was

locked (3 mL) for 15 min/day, and was flushed with 10 mL of 0.9% normal saline. In the

outpatient setting, ALT was instilled by the nursing staff once weekly. In documented

bacteremia cases or upon initiation of empiric therapy, catheter hub cultures were performed.

Only endoluminal CLABSI was considered to be an attributable endpoint for infection in this

study. Ethanol was used in 266 CVCs and 0.9% normal saline was used in 222 CVCs. In total,

10 infections over 14 262 catheter-days (0.70/1000 catheter-days) were documented in the

ethanol lock group, compared with 16 infections over 13 483 catheter-days (1.19/1000 CVC-

days) in the normal saline group. The most common pathogen cultured was CoNS. The reduction

in infection rate of 41% was not statistically significant (P=0.19). One ethanol-treated patient

experienced syncope after flushing through the first lock solution. No life-threatening adverse

events were observed [22]. Compatibility and stability studies have been conducted using

various ethanol concentrations and combinations. A lock solution of 70% ethanol and 10

Page 10 of 38
units/mL of heparin sodium has been shown to maintain stability at 24, 48 and 72 h [40].

Concerns with ethanol use in lock solutions are discussed further below.

4. Prophylactic studies in pediatric cancer patients

CLABSIs are the leading complication of pediatric oncology patients with CVCs [7]. CLABSI

rates ranging from 1.6 to 2.8 episodes per 1000 patient-days have been reported, leading to

clinical sepsis, admission to intensive care, catheter removal and delay of chemotherapy

administration [41]. Neonates in the neonatal intensive care unit can undergo an average of 3.2

catheter manipulations per day (range 0–15) [41]. Antimicrobial lock solutions containing VH

(n=3), mEDTA (n=1), ethanol (n=1), taurolidine (n=3) and combination solutions (vancomycin

plus amikacin and vancomycin plus ciprofloxacin) have been evaluated for prevention of

CLABSI in the pediatric oncology setting (Table 2).

4.1. Ethanol-based lock solutions

Ethanol at 70% concentration in lock solutions has led to a reduction in CLABSI rates ranging

from 42% to 80% in children [7]. Schoot et al. conducted the first randomized, double-blind,

multi-center trial evaluating ethanol locks specifically in pediatric oncology patients [23].

Patients were randomized to receive an ethanol lock (70%, n=153) or a heparin lock (100

units/mL, n=154) for a 2-h single dwell. After 2 h, a normal saline flush was performed and the

CVC was locked with heparin. CVC-associated infections were defined in accordance with

national guidelines, and patients had to have accompanying signs or symptoms. The CLABSI

rate was 10% (16/153) in the ethanol arm and 19% (29/154) in the heparin arm (hazard rate 0.53,

95% confidence interval 0.29–0.98). The incidence of CLABSI was 0.77 and 1.46 per 1000

Page 11 of 38
catheter-days (P=0.039) in the ethanol and heparin arms, respectively. The most common

pathogens were Gram-positive (64%), Gram-negative (16%), polymicrobial (13%) and fungi

(7%). Fewer CLABSIs were caused by Gram-positive organisms in the ethanol group (P=0.012).

There were no significant differences in the number of positive blood cultures between groups

[23]. Five CVCs were removed in the ethanol group vs. 12 in the heparin group due to CLABSIs

(P=0.077). More patients reported side effects in the ethanol group including nausea, taste

alteration, dizziness and blushing. All symptoms were transient and limited to the time of

flushing [23]. The authors concluded that ethanol did not contribute to CVC occlusion or CVC-

related symptomatic venous thromboembolism (VTE).

Concerns about catheter occlusion have been suggested but not confirmed with ethanol locks

[42]. Heparin has been shown to precipitate at higher ethanol concentrations [7]. Ethanol (50–

70%) plus heparin solutions have been shown to be stable at room temperature in polypropylene

syringes for at least 14–28 days, and dwell times ranging from 15 min to 48 h have been studied

[43,44]. Additional concerns with ethanol-based lock solutions, specifically in pediatrics, are

discussed in detail below.

4.2. Vancomycin-based lock solutions

Schwartz et al. conducted a randomized, double-blind trial to investigate the impact of

prophylactic VH flush solution on the incidence of CVC-related bacteremia secondary to

colonization of the catheter lumen [24]. They evaluated outcomes related to vancomycin-

susceptible organisms [24]. Patients were included if they were under 20 years of age with

tunneled central venous catheters in place, had cancer-related neutropenia, or were receiving

Page 12 of 38
active chemotherapy. Study subjects were randomized to receive either heparin 10 units/mL

alone (n=24) or vancomycin 25 µg/mL plus heparin 9.57 units/mL (n=21). Bacteremia

attributable to intraluminal colonization was defined as a peripheral blood sample with <10% of

the colony count from a concurrent sample from the CVC, without signs of an exit-site infection

due to the same organism. This definition was consistent with the literature at the time of the

study [45]. The catheter was flushed with 5 mL of solution, allowed to dwell in the line daily and

was continued until the catheter was removed. For neutropenic patients presenting with fever,

swabs were taken of the catheter hubs and blood cultures were drawn prior to administration of

empiric intravenous antibiotics. Of the 21 patients who received the VH solution, one patient

(4.8%) was diagnosed with bacteremia (Klebsiella pneumoniae) attributed to the luminal

colonization of the CVC, compared with six infections (25%) in the heparin group (P=0.035).

Five of the six infections were secondary to CoNS, and one was a polymicrobial infection with

E. coli and CoNS. The VH arm was associated with significantly longer CVC survival without

colonization with a vancomycin-susceptible organism (P=0.04). The investigators also found that

despite the additional cost of preparation and ingredients for VH, there was a significant cost

saving. The use of VH lock therapy decreased the use of additional systemic antibiotic therapy

required for the treatment of CLABSI. The cost of hospitalization of CLABSI was found to be

approximately $24 000 for each case, providing evidence of significant cost-savings when VH

lock solution was utilized [24].

Rackoff et al. conducted a randomized, controlled trial investigating the use of VH in children

with and without cancer using an indwelling cuffed CVC [25]. Similar to Schwartz et al., the

primary objective of the study was to determine whether the addition of vancomycin to flush

Page 13 of 38
solution would significantly reduce the incidence of bacteremia attributable to luminal

colonization due to vancomycin-susceptible organisms. There were 55 children with cancer and

eight children receiving total parenteral nutrition with short bowel syndrome, totaling 63

patients. Patients were randomized to receive either vancomycin 25 µg/mL plus heparin 100

units/mL (n=32) or heparin 100 units/mL alone (n=31). Thrice-weekly dressing changes and a

maximum of four line interruptions in a 24-h period were employed. Catheters were flushed with

VH solution (3 mL) from a multi-use vial or a single-dose heparin syringe. Prophylactic

trimethoprim/sulfamethoxazole was given to patients with acute lymphoblastic leukemia and

acute myeloid leukemia. The definition of bacteremia attributable to luminal colonization was

consistent with Schwartz et al. and prevailing evidence at the time [24,45]. Among the 63

patients, calculated overall CLABSI rates were 9.7% (three of 31) and 37.5% (12 of 32) in the

heparin and VH arms, respectively (no P-value available). Rates of bacteremia attributable to

luminal colonization with vancomycin-susceptible organisms alone was 1.4/1000 CVC-days in

the VH arm and 0.6/1000 CVC-days in the heparin-alone group (P=0.25). Rates of bacteremia

secondary to Gram-negative organisms and yeasts were seven of 32 (21.9%) and one of 31

(3.2%) in the VH and heparin groups, respectively (P=0.03). Common pathogens identified were

viridans group Streptococcus and Enterococcus spp. No vancomycin resistance was seen in this

study. No adverse events were mentioned. The authors concluded that there were no benefits of

the VH lock solution arm, and noted higher rates of infection due to Gram-negative bacteria or

yeast [25].

A double-blind, randomized trial conducted by Barriga et al. evaluated VH lock therapy for

prevention of bacteremia in neutropenic and non-neutropenic pediatric patients [26]. The study

Page 14 of 38
mainly included children, but also included six patients aged >20 years. Study subjects were

randomized to receive heparin 25 units/mL lock solution (n=44) or vancomycin 25 µg/mL plus

heparin 25 units/mL (n=39). During catheter placement, 5 mL of lock solution was flushed daily.

Bacteremia was defined as any positive blood culture, central or peripheral, during the initial

febrile episode following CVC placement. Bacteremia was detected in 44 patients: 26 (59%) in

the heparin group and 18 (46.1%) in the VH group. Of those 44 instances, 24 cases were positive

cultures from the CVC alone and 13 were CVC plus peripheral. The infection rate was found to

be 2.25 per 1000 CVC-days in the VH arm compared with 3.0 per 1000 CVC-days for heparin

alone [26]. Infections secondary to vancomycin-susceptible organisms (VSOs) only occurred in

16 of 39 (41%) in the heparin group and seven of 44 (15.9%) in the VH group (P=0.19).

Infections from non-neutropenic patients, including bacteremia secondary to VSOs, occurred in

nine patients in the heparin group compared with no occurrences in the VH group (P=0.01). No

further data by site-specific cultures were available and no reports of adverse events were noted.

S. epidermidis and E. coli were the most common organisms isolated.

4.3. EDTA + minocycline

Early investigation of tetracycline-based lock solutions originated from animal models and in-

vitro models of CVC infection. The addition of EDTA to lock solutions disrupts intact biofilms

and provides synergy with antimicrobials against slime-producing organisms [46].

Chatzinikolaou et al. investigated the use of mEDTA lock solution in pediatric cancer patients

with newly placed implantable ports. Forty-eight patients who had monthly heparin port flushes

were followed for 6 months [27]. The investigators prospectively followed 14 children by

locking 2 mL of minocycline 3 mg/mL plus EDTA 30 mg/mL into the ports and remained in situ

Page 15 of 38
weekly. Patients were followed for 6 months, until ports were removed or death, whichever

occurred first. Patients were evaluated for signs and symptoms of infections and mEDTA-related

adverse events. A ‘probable’ CRBSI was defined as a positive peripheral blood culture with an

organism typically associated with CRBSI with systemic symptoms and no apparent source other

than the CVC. A ‘definite’ CRBSI included the above conditions plus isolation of the same

organism centrally. No infections were found in the patients receiving mEDTA, but 10 of 48

(20.8%) infections (including definite, probable or local CVC infections) were in the heparin

group (P=0.06). Nine patients in the heparin group required CVC removal and two patients had

noted thrombosis. The mEDTA group was found to have an infection rate of 0 per 1000 catheter-

days compared with 2.23/1000 catheter-days in the heparin-alone group (P=0.05) [27].The use of

mEDTA did not result in thrombotic events or other adverse events. EDTA combination lock

solutions have been studied elsewhere in tetracycline-based solutions with no reports of

incompatibility [38].

4.4. Combination antimicrobial lock solutions

Henrickson et al. conducted a randomized double-blind controlled study in immunocompromised

children to determine the efficacy of vancomycin, ciprofloxacin and heparin lock solutions, both

alone and in combination, as prophylaxis for CVC infections [29]. Children were all under the

age of 20 years, and the cause of immunosuppression was primarily secondary to cancer.

Endpoints included possible, probable and definite catheter infections and were determined by

cultures and temperature. Definite CVC-related infections required confirmed positive cultures

from CVC and peripheral specimens, quantitatively 10-fold higher from the CVC. Probable and

possible CVC-related infections were less stringent, with total absence of qualitative cultures or

Page 16 of 38
peripheral cultures. Patients were randomized to receive one of the following lock solutions:

heparin 9.73 units/mL (n=80), vancomycin 25 µg/mL plus heparin 9.73 units/mL (n=35), and

vancomycin 25 µg/mL plus ciprofloxacin 2 µg/mL plus heparin 9.73 units/mL in combination

(n=38) (VCH). Infection rates were 12 of 80 (15%) for the heparin-alone group, one of 35

(2.9%) for the VH group, and one of 38 (2.6%) for the VCH group. The infection rate for the

heparin-alone group was 0.67/1000 catheter-days. There was a significant reduction in infection

rates per 1000 CVC-days in the VH group (0.13/1000 catheter-days; P=0.06) and VCH group

(0.09/1000 catheter-days; p=0.03) when compared with the heparin-alone group. The most

common pathogens identified were CoNS (n=8), Acinetobacter spp. (n=2) and E. coli (n=2) [29].

No specific adverse events were highlighted, and no reports of incompatibility were reported in

the combination antimicrobial lock solutions.

In a randomized, placebo-controlled, double-blind trial, Daghistani et al. investigated the use of

an aminoglycoside plus vancomycin lock solution in pediatric cancer patients [28]. Patients

under the age of 22 years with a malignancy diagnosis and indwelling catheter were eligible for

inclusion. Thirty-two of 64 patients had acute lymphoblastic leukemia, with the remaining

patients having various tumor types or Hodgkin’s lymphoma. Patients were randomized to

receive either standard heparin 100 units/mL (n=33) solution or the study solution, vancomycin

25 µg/mL plus amikacin 25 µg/mL plus heparin 100 units/mL (VAH) (n=31; 28 patients blinded

and three unblinded). Catheters were flushed with 5 mL of solution each time the line was

utilized for a total of 17 months. Cultures were obtained from patients presenting with fever.

Bacteremia was defined as a positive blood culture from the CVC and/or peripheral sample. The

heparin arm resulted in three cases of cellulitis and three cases of bacteremia, compared with two

Page 17 of 38
cases of cellulitis and two cases of bacteremia in the VAH arm. The rate of CVC-related

infection for the standard solution (heparin) was 0.3/1000 catheter-days and 0.2/1000 catheter-

days for the VAH solution, which was not statistically different (no P-value reported). The two

cases of bacteremia in the study arm were due to S. epidermidis and Stenotrophomonas

maltophilia, while Candida spp. (n=2) and S. aureus were responsible for bacteremia in the

heparin group [28]. No adverse events were highlighted in the study. Visual confirmation of

physical compatibility of the VAH solution was noted by the authors.

4.5. Taurolidine

Handrup et al. conducted a prospective, randomized, controlled, open-label study evaluating the

use of taurolidine lock therapy vs. heparin lock alone [32]. The study sought to determine if

taurolidine lock therapy could reduce the number of CRBSIs in 113 pediatric oncology patients

(130 placed tunneled CVCs). In the experimental arm, catheters were locked with 2.5 mL of

taurolidine 1.35%, sodium citrate 4% and heparin 100 units/mL. In the standard arm, catheters

were locked with 250 units of heparin in 2.5 mL of sterile normal saline. CRBSI was defined in

accordance with the 2009 IDSA guidelines (e.g. positive blood samples from central and

peripheral samples, with >2 h earlier growth in the CVC sample) [16]. A lower rate of CRBSI

was observed in the taurolidine group compared with the heparin group: 0.4 per 1000 CVC-days

vs. 1.4 per 1000 CVC-days, respectively (95% confidence interval 0.09–0.61; P=0.001).

Taurolidine was also shown to reduce the number of total BSIs (from 40% to 10.9%) and time to

first CRBSI. However, the study did not show a significant difference in CVC survival (median

236 and 300 days in taurolidine and heparin groups, respectively; P=0.19). The most common

adverse event was a brief, unpleasant taste in the mouth during catheter flush in the taurolidine

Page 18 of 38
group. Rates of mechanical complications (thrombosis, misplacement or leak) were minimal and

similar in both groups (6.3% and 6.2% in experimental and control arm, respectively); neither

study arm required CVC removal due to thrombus formation. The most common pathogen

identified was S. epidermidis (n=2) [32].

Dumichen et al. conducted a randomized controlled trial to compare taurolidine citrate with

heparin in 71 pediatric patients aged 1–18 years with hematological malignancies [31]. Patients

were randomized to either taurolidine 1.35% and sodium citrate 4% (n=35) or heparin 100

units/mL (n=36). The numbers of patients treated in the bone marrow transplant unit and

oncology wards were similar between groups. Patients received sulfamethoxazole/trimethoprim

for Pneumocystis jirovecii pneumonia prophylaxis and were examined daily for fever and signs

of infection. The primary endpoint was CVC colonization, while the numbers of BSIs and

thrombotic events were secondary endpoints. Approximately 25% (13 of 51) of CVCs showed

colonization, four of which were considered significant colonization (two in the heparin group

and two in the taurolidine group). Two of the 35 patients (5%) in the taurolidine group

experienced a BSI compared with nine of the 36 patients (25%) in the heparin group. The rate of

CLABSIs was significantly lower in the taurolidine group: 0.3 per 1000 catheter-days compared

with 1.3 per 1000 catheter-days in the heparin group (P=0.03). Catheter occlusion due to a

confirmed thrombus occurred in three patients in the taurolidine arm and two patients in the

heparin arm [31]. Seven patients (20%) experienced side effects in the taurolidine arm, including

discomfort in the chest and neck, peri-oral dysaesthesia, abnormal taste sensations, nausea and

vomiting; three of these patients discontinued therapy [31].

Page 19 of 38
Simon et al. conducted a randomized, double-blind study comparing taurolidine citrate with

heparin lock solution in a single tertiary care center [30]. Patients had an average age of 8 years

and a malignancy diagnosis requiring CVC insertion for treatment. The primary endpoint was

BSI secondary to CoNS. In accordance with German guidelines, a BSI was defined as positive

cultures from the CVC in a patient with a clinical syndrome suggestive of infection [47]. In total,

179 patients were included (heparin, n=90; taurolidine, n=89). Patients were randomized to

receive heparin 100 units/mL in 2 mL of normal saline or taurolidine 1.35% plus sodium citrate

4%. Ports were locked one or two times weekly by clinic or in-home staff members. There was

no difference in the overall rates of BSI between groups [27% (24 of 90) in the heparin group vs.

24% (21 of 89) in the taruolidine group; P=0.74]. In comparing patients with CoNS alone, there

was a significant reduction in the taurolidine group (11%) compared with the heparin group

(47%) (P=0.004). The rates of infection per 1000 CVC-days were not different for overall BSI,

but were significantly lower for BSI secondary to CoNS (2.30/1000 days vs. 0.45/1000 days;

P=0.004) in the taurolidine group. An abnormal taste following flushing of the taurolidine lock

was documented. No patients discontinued the study lock due to an adverse event.

Taurolidine has been evaluated for compatibility and stability in combination with trisodium

citrate and heparin; as a lock solution, no incompatibilities have been reported and stability has

been observed in hemodialysis patients for an average of 30.5 days [38,48]. This combination is

available under the brand name ‘Taurolock’ in Europe [49]. Liu et al. published a systematic

review and meta-analysis of prophylactic taurolidine lock therapy in various patient populations

beyond the scope of this review in cancer patients [50].

Page 20 of 38
6. Discussion

Cancer patients require prolonged central venous access that can ultimately lead to bacterial

colonization of the CVC and BSIs. Prophylactic ALT may offer clinicians a method to prevent

bacterial colonization, reducing the rates of CLABSI and thus prolonging CVC survival. ALT

has been shown to be highly effective in other patient populations, including haemodialysis-

dependent patients, for preventing subsequent CLABSI and exposure to systemic antibiotics

[48]. The impact of CLABSI has been projected to increase the cost by US$4,888–29,000

[51,52]. By preventing CLABSI, the need for subsequent hospitalization – and thus systemic

antibiotics, catheter removal and replacement – is avoided or reduced.

Despite the potential inherent advantages of ALT in cancer patients, only three studies in adults

and 10 studies in children that met the search criteria were identified and therefore included in

this review. Ethanol-based solutions (70%) were used in two of the studies, while vancomycin

plus heparin was used in a single study. Five of the pediatric studies utilized vancomycin-based

solutions, two of which used additional combinations including amikacin and ciprofloxacin. The

additional studies among children utilized ethanol, mEDTA and taurolidine-based lock solutions.

Vancomycin is the most commonly utilized antibiotic in lock solutions, primarily due to the

etiology of CLABSI being primarily Gram-positive organisms [38]. Results among studies of

vancomycin-based solutions were variable; the single adult study [21] demonstrated a significant

reduction in CLABSI rates, while only two of the five pediatric studies [24,29] demonstrated a

similar positive impact. The impact in the study by Henrickson et al. [29] was primarily seen in

those receiving vancomycin plus ciprofloxacin in solution. The concentration of vancomycin

Page 21 of 38
used in all studies (25 µg/mL) is significantly lower than that used in most reports outside of

prophylaxis in cancer patients, given the elevated minimum inhibitory concentration associated

with bacterial biofilms. However, it is unknown if higher concentrations in a prophylactic

modality would be useful. Dwell times varied from 1 h [21] to daily in the majority of studies. It

was unclear how much interruption of ALT occurred in most studies. Extended dwell times

taking advantage of prolonged stability may be useful, especially in patients receiving multiple

cycles of chemotherapy or those being treated outside of an inpatient setting. The optimal dwell

time is unknown; however, shorter dwell times in non-vancomycin-based solutions have been

associated with lower effectiveness. Compatibility and stability of vancomycin-based lock

solutions is discussed in detail elsewhere [38,53]. Widespread use of prophylactic antibiotic-

based lock solutions, including vancomycin, may contribute to development of resistance. The

limited results and variability in study quality does not allow for firm conclusions or

recommendations to be made for vancomycin-based solutions in a prophylactic modality.

The two studies investigating ethanol-based solutions in adult cancer patients provided differing

results. Sanders et al. demonstrated a significant reduction in CLABSI compared with heparin,

although the sample size was relatively small (n=60) [8]; the infection rates per 1000 CVC-days

were relatively low in both study arms. In a much larger study of 448 patients, primarily with

hematological tumors, Slobbe et al. did not show a significant reduction in infection rates,

although there was a 41% decrease in CLABSI from 1.19 to 0.7 per 1000 CVC-days [22].

The single study of ethanol-based lock solution in children [23] demonstrated a significant

reduction in CLABSI rates with only a single dwell of 2 h. Dwell times in the two adult studies

Page 22 of 38
varied, although Slobbe et al. only used a 15-min dwell time [22]. Short dwell times in vitro have

been associated with rebound growth, and recently published prospective studies in the adult

intensive care setting have not shown a reduction in CLASBI when using <2 h dwell times.

Further investigation is needed to determine the optimal dwell time.

Serious adverse events are of concern with ethanol-based lock solutions. Slobbe et al. reported

syncope in one patient following flushing of the initial ethanol lock solution, but no additional

adverse events were reported [22]. Only mild effects including nausea, vomiting and dizziness

were reported by Schoot et al. among children [23]. Aspiration of such solutions is preferred and

may mitigate adverse effects, which may be exaggerated in those of younger age and lower body

weight. Despite the potential effects, no cases of serious symptomatic intoxication from ethanol

locks have been reported. Pediatric catheters have a luminal volume of less than 1 mL, which is

equivalent to only 5% of a standard alcoholic drink [7]. The American Academy of Pediatrics

Committee on Drugs has recommended that the amount of ethanol contained in any preparation

should not be able to produce a blood alcohol concentration (BAC) >25 mg/dL after a single

dose [54]. The maximum predicted BAC for even a 5-kg infant receiving an intravenous bolus of

1.5 mL of 70% ethanol is 0.024% [7]. Therefore, concerns of systemic exposure based on this

data should be minimal, especially if aspiration is utilized as recommended. Catheter occlusions

and increased risk of VTE are suspected in ethanol-based lock solutions. There are some

concerns regarding compatibility issues with heparin depending on the ethanol concentration

utilized. These studies did not use an anticoagulant in solution, and no attributable increased risk

of occlusions or VTE were made by the authors. Ethanol may interfere with the plastic (PVC)

polymers in some standardized catheters, degrading the plastic over time, although this does

Page 23 of 38
appear to be CVC-dependent [22]. Attention to CVC type is important when using ethanol for a

lock solution, especially for prolonged dwells. Despite some of these issues, there are inherent

advantages of ethanol-based lock solutions, including broad antibacterial and antifungal activity,

low potential for induction of resistance and relatively low cost [39,53].

The antiseptic agent, taurolidine, was found to be effective in two [31,32] of the three pediatric

studies, reducing CLABSI rates compared with heparin solutions. The combinations used in

these studies of taurolidine, sodium citrate and heparin are marketed under the branded product,

‘TauroLock’. The addition of citrate offers a potential antibiofilm and modest antibacterial

combination. Taurolidine dwells varied in the studies, although the stability of the combination

used has been shown to be greater than 28 days, lending to long dwells between office visits or

outpatient chemotherapy. Taurolidine was associated with several adverse events upon flushing,

including abnormal taste, nausea, vomiting, chest and neck discomfort, and peri-oral

dysaesthesia. In one study, three patients discontinued the taurolidine lock due to adverse events.

As mentioned previously with ethanol-based solutions, aspiration of the lock solution is

preferred to reduce the risk of these adverse events. While not available for commercial use in

many parts of the world, taurolidine-based lock solutions offer many of the inherent advantages

seen with ethanol-based solutions, while avoiding the need for an antibiotic-based solution.

Providers should consider the overall limitations of ALT. In general, prophylactic ALT may not

be useful in each cancer patient or in broad application for an entire cancer type. This decision

should be based on local data including, but not limited to, CLABSI rates, epidemiology, other

infection control and CVC practices, and cancer populations. While lock solutions dwell in the

Page 24 of 38
catheter lumen, the line cannot be utilized until it is aspirated or flushed. The dwell time and

frequency of dwells will need to be considered in order to maintain convenience of the line.

Oncology patients often practise catheter self-care in an outpatient setting due to the long-term

nature of chemotherapy, so use of prophylactic ALT may require patient education to ensure

continuity to outpatient settings. The question of development of resistance to antimicrobial

agents employed in lock solutions is one that has yet to be answered; the studies included in this

review were not designed to answer this question. Other concerns include the documentation of

lock solution administration and removal, as well as preparation methods, stability and

compatibility of solutions. Stability and compatibility of lock solutions have been evaluated for

practicality. The use of non-antibiotic-based solutions in a prophylactic modality should be given

strong consideration in the future. Proper education of nursing staff and patients on aspiration of

solutions and potential adverse events, especially when using ethanol- or taurolidine-based

solutions, is needed. Development of ALT institutional guidelines in consultation with infectious

disease experts is preferred to avoid proper utilization and unnecessary adverse events.

7. Conclusion

Data for prophylactic ALT in adult cancer patients are lacking. While the data in pediatric cancer

patients are more robust, the quality of evidence varies significantly as do the lock solutions,

dwell times and outcome assessments utilized. The majority of data, especially studies using

antiseptics such as ethanol and taurolidine, demonstrate a reduction in CLABSI rate; however,

associated adverse events when the solution is flushed (as opposed to aspirated) do occur, but

appear to be self-limiting. No firm recommendations can be made for use of antibiotic-based

Page 25 of 38
regimens including vancomycin-based solutions. Use of ALT does not replace proper infection

control measures, aseptic technique and routine CVC care. Prophylactic ALT may be indicated

in targeted cancer patients or cancer populations at institutions with high rates of CLABSI

despite infection control and CVC care measures.

Funding: None.

Competing interests: None declared.

Ethical approval: Not required.

References

[1] O'Grady NP, Alexander M, Burns LA, Dellinger EP, Garland J, Heard SO, et al. Guidelines

for the prevention of intravascular catheter-related infections. Am J Infect Control 2011;39:S1–

34.

[2] Centers for Disease Control and Prevention. Vital signs: central line-associated blood stream

infections – United States, 2001, 2008, and 2009. MMWR 2011;60:243–8.

[3] Wisplinghoff H, Bischoff T, Tallent SM, Seifert H, Wenzel RP, Edmond MB. Nosocomial

bloodstream infections in US hospitals: analysis of 24,179 cases from a prospective nationwide

surveillance study. Clin Infect Dis 2004;39:309–17.

[4] Butt T, Afzal RK, Ahmad RN, Salman M, Mahmood A, Anwar M. Bloodstream infections in

febrile neutropenic patients: bacterial spectrum and antimicrobial susceptibility pattern. J Ayub

Med Coll Abbottabad 2004;16:18–22.

Page 26 of 38
[5] Celebi S, Sezgin ME, Cakir D, Baytan B, Demirkaya M, Sevinir B, et al. Catheter-associated

bloodstream infections in pediatric hematology-oncology patients. Pediatr Hematol Oncol

2013;30:187–94.

[6] Crnich CJ. Estimating excess length of stay due to central line-associated bloodstream

infection: separating the wheat from the chaff. Infect Control Hosp Epidemiol 2010;31:1115–7.

[7] Wolf J, Shenep JL, Clifford V, Curtis N, Flynn PM. Ethanol lock therapy in pediatric

hematology and oncology. Pediatr Blood Cancer 2013;60:18–25.

[8] Sanders J, Pithie A, Ganly P, Surgenor L, Wilson R, Merriman E, et al. A prospective

double-blind randomized trial comparing intraluminal ethanol with heparinized saline for the

prevention of catheter-associated bloodstream infection in immunosuppressed haematology

patients. J Antimicrob Chemother 2008;62:809–15.

[9] Miller A. Hospital reporting and ‘never events’. Medicare Pat Manag 2009;4:20–2.

[10] Scott RD. The direct medical costs of healthcare-associated infections in US hospitals and

the benefits of prevention. Atlanta, GA: Division of Healthcare Quality Promotion, Centers for

Disease Control and Prevention; 2009. [Link]

[accessed 9 November 2015].

[11] van de Wetering MD, van Woensel JB. Prophylactic antibiotics for preventing early central

venous catheter Gram positive infections in oncology [Link] Database System Rev

2007;1:CD003295.

[12] Butt T, Afzal RK, Ahmad RN, Hussain I, Anwar M. Central venous catheter-related

bloodstream infections in cancer patients. J Coll Phys Surg Pakistan 2004;14:549–52.

[13] Manierski C, Besarab A. Antimicrobial locks: putting the lock on catheter infections. Adv

Chron Kid Dis 2006;13:245–58.

Page 27 of 38
[14] Bookstaver PB, Williamson JC, Tucker BK, Raad II, Sherertz RJ. Activity of novel

antibiotic lock solutions in a model against isolates of catheter-related bloodstream infections.

Ann Pharmacother 2009;43:210–9.

[15] Safdar N, Maki DG. Use of vancomycin-containing lock or flush solutions for prevention of

bloodstream infection associated with central venous access devices: a meta-analysis of

prospective, randomized trials. Clin Infect Dis 2006;43:474–84.

[16] Mermel LA, Allon M, Bouza E, Craven DE, Flynn P, O'Grady NP, et al. Clinical practice

guidelines for the diagnosis and management of intravascular catheter-related infection: 2009

update by the Infectious Diseases Society of America. Clin Infect Dis 2009;49:1–45.

[17] National Comprehensive Cancer Network. National Comprehensive Cancer Network

guidelines for the management of prostate cancer. Fort Washington, PA: NCCN; 2015.

[18] Schiffer CA, Mangu PB, Wade JC, Camp-Sorrell D, Cope DG, El-Rayes BF, et al. Central

venous catheter care for the patient with cancer: American Society of Clinical Oncology clinical

practice guideline. J Clin Oncol 2013;31:1357–70.

[19] Moher D, Shamseer L, Clarke M, Ghersi D, Liberati A, Petticrew M, et al. Preferred

reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015 statement.

System Rev 2015;4:1.

[20] Andrews JC, Schunemann HJ, Oxman AD, Pottie K, Meerpohl JJ, Coello PA, et al.

GRADE guidelines: 15. Going from evidence to recommendation – determinants of a

recommendation's direction and strength. J Clin Epidemiol 2013;66:726–35.

[21] Carratala J, Niubo J, Fernandez-Sevilla A, Juve E, Castellsague X, Berlanga J, et al.

Randomized, double-blind trial of an antibiotic-lock technique for prevention of Gram-positive

Page 28 of 38
central venous catheter-related infection in neutropenic patients with cancer. Antimicrob Agents

Chemother 1999;43:2200–4.

[22] Slobbe L, Doorduijn JK, Lugtenburg PJ, El Barzouhi A, Boersma E, van Leeuwen WB, et

al. Prevention of catheter-related bacteremia with a daily ethanol lock in patients with tunnelled

catheters: a randomized, placebo-controlled trial. PloS One 2010;5:e10840.

[23] Schoot RA, van Ommen CH, Stijnen T, Tissing WJ, Michiels E, Abbink FC, et al.

Prevention of central venous catheter-associated bloodstream infections in paediatric oncology

patients using 70% ethanol locks: a randomised controlled multi-centre trial. Eur J Cancer

2015;51:2031–8.

[24] Schwartz C, Henrickson KJ, Roghmann K, Powell K. Prevention of bacteremia attributed to

luminal colonization of tunneled central venous catheters with vancomycin-susceptible

organisms. J Clin Oncol 1990;8:1591–7.

[25] Rackoff WR, Weiman M, Jakobowski D, Hirschl R, Stallings V, Bilodeau J, et al. A

randomized, controlled trial of the efficacy of a heparin and vancomycin solution in preventing

central venous catheter infections in children. J Pediatr 1995;127:147–51.

[26] Barriga FJ, Varas M, Potin M, Sapunar F, Rojo H, Martinez A, et al. Efficacy of a

vancomycin solution to prevent bacteremia associated with an indwelling central venous catheter

in neutropenic and non-neutropenic cancer patients. Med Pediatr Oncol 1997;28:196–200.

[27] Chatzinikolaou I, Zipf TF, Hanna H, Umphrey J, Roberts WM, Sherertz R, et al.

Minocycline-ethylenediaminetetraacetate lock solution for the prevention of implantable port

infections in children with cancer. Clin Infect Dis 2003;36:116–9.

[28] Daghistani D, Horn M, Rodriguez Z, Schoenike S, Toledano S. Prevention of indwelling

central venous catheter sepsis. Med Pediatr Oncol 1996;26:405–8.

Page 29 of 38
[29] Henrickson KJ, Axtell RA, Hoover SM, Kuhn SM, Pritchett J, Kehl SC, et al. Prevention of

central venous catheter-related infections and thrombotic events in immunocompromised

children by the use of vancomycin/ciprofloxacin/heparin flush solution: a randomized,

multicenter, double-blind trial. J Clin Oncol 2000;18:1269–78.

[30] Simon A, Ammann RA, Wiszniewsky G, Bode U, Fleischhack G, Besuden MM.

Taurolidine-citrate lock solution (TauroLock) significantly reduces CVAD-associated

grampositive infections in pediatric cancer patients. BMC Infect Dis 2008;8:102.

[31] Dumichen MJ, Seeger K, Lode HN, Kuhl JS, Ebell W, Degenhardt P, et al. Randomized

controlled trial of taurolidine citrate versus heparin as catheter lock solution in paediatric patients

with haematological malignancies. J Hosp Infect 2012;80:304–9.

[32] Handrup MM, Moller JK, Schroder H. Central venous catheters and catheter locks in

children with cancer: a prospective randomized trial of taurolidine versus heparin. Pediatr Blood

Cancer 2013;60:1292–8.

[33] Handrup MM, Fuursted K, Funch P, Moller JK, Schroder H. Biofilm formation in long-term

central venous catheters in children with cancer: a randomized controlled open-labelled trial of

taurolidine versus heparin. APMIS 2012;120:794–801.

[34] Rubie H, Juricic M, Claeyssens S, Krimou A, Lemozy J, Izard P, et al. Morbidity using

subcutaneous ports and efficacy of vancomycin flushing in cancer. Arch Dis Child 1995;72:325–

9.

[35] Steinberg JP, Robichaux C, Tejedor SC, Reyes MD, Jacob JT. Distribution of pathogens in

central line-associated bloodstream infections among patients with and without neutropenia

following chemotherapy: evidence for a proposed modification to the current surveillance

definition. Infect Control Hosp Epidemiol 2013;34:171–5.

Page 30 of 38
[36] Dudeck MA, Weiner LM, Allen-Bridson K, Malpiedi PJ, Peterson KD, Pollock DA, et al.

National Healthcare Safety Network (NHSN) report, data summary for 2012, Device-associated

module. Am J Infect Control 2013;41:1148–66.

[37] Raad I, Buzaid A, Rhyne J, Hachem R, Darouiche R, Safar H, et al. Minocycline and

ethylenediaminetetraacetate for the prevention of recurrent vascular catheter infections. Clin

Infect Dis 1997;25:149–51.

[38] Bookstaver PB, Rokas KE, Norris LB, Edwards JM, Sherertz RJ. Stability and compatibility

of antimicrobial lock solutions. Am J Health Syst Pharm 2013;70:2185–98.

[39] Maiefski M, Rupp ME, Hermsen ED. Ethanol lock technique: review of the literature. Infect

Control Hosp Epidemiol 2009;30:1096–108.

[40] Takla TA, Zelenitsky SA, Vercaigne LM. Effect of ethanol/trisodium citrate lock on

microorganisms causing hemodialysis catheter-related infections. J Vasc Access 2007;8:262–7.

[41] Mahieu LM, De Dooy JJ, Lenaerts AE, Ieven MM, De Muynck AO. Catheter manipulations

and the risk of catheter-associated bloodstream infection in neonatal intensive care unit patients.

J Hosp Infect 2001;48:20–6.

[42] Kayton ML, Garmey EG, Ishill NM, Cheung NK, Kushner BH, Kramer K, et al.

Preliminary results of a phase I trial of prophylactic ethanol-lock administration to prevent

mediport catheter-related bloodstream infections. J Pediatr Surg 2010;45:1961–6.

[43] Cober MP, Johnson CE. Stability of 70% alcohol solutions in polypropylene syringes for

use in ethanol-lock therapy. Am J Health Syst Pharm 2007;64:2480–2.

[44] Pomplun M, Johnson JJ, Johnston S, Kolesar JM. Stability of a heparin-free 50% ethanol

lock solution for central venous catheters. J Oncol Pharm Pract 2007;13:33–7.

Page 31 of 38
[45] Raucher HS, Hyatt AC, Barzilai A, Harris MB, Weiner MA, LeLeiko NS, et al. Quantitative

blood cultures in the evaluation of septicemia in children with Broviac catheters. J Pediatr

1984;104:29–33.

[46] Raad II, Hachem RY, Hanna HA, Fang X, Jiang Y, Dvorak T, et al. Role of ethylene

diamine tetra-acetic acid (EDTA) in catheter lock solutions: EDTA enhances the antifungal

activity of amphotericin B lipid complex against candida embedded in biofilm. Int J Antimicrob

Agents 2008;32:515–8.

[47] Beutel K, Simon A. [Diagnostics and management of central venous line infections in

pediatric cancer patients]. Klinische Padiatrie 2005;217(Suppl. 1):S91–100.

[48] Zwiech R, Adelt M, Chrul S. A taurolidine-citrate-heparin lock solution effectively

eradicates pathogens from the catheter biofilm in hemodialysis patients. Am J Ther

2016;23:e363–8.

[49] [Link] [accessed 9 December 2016].

[50] Liu Y, Zhang AQ, Cao L, Xia HT, Ma JJ. Taurolidine lock solutions for the prevention of

catheter-related bloodstream infections: a systematic review and meta-analysis of randomized

controlled trials. PloS One 2013;8:e79417.

[51] Rosenthal VD, Guzman S, Migone O, Crnich CJ. The attributable cost, length of hospital

stay, and mortality of central line-associated bloodstream infection in intensive care departments

in Argentina: a prospective, matched analysis. Am J Infect Control 2003;31:475–80.

[52] Pittet D, Tarara D, Wenzel RP. Nosocomial bloodstream infection in critically ill patients.

Excess length of stay, extra costs, and attributable mortality. JAMA 1994;271:1598–601.

[53] Justo JA, Bookstaver PB. Antibiotic lock therapy: review of technique and logistical

challenges. Infect Drug Resist 2014;7:343–63.

Page 32 of 38
 [54] Committee on Drugs. Ethanol in liquid preparations intended for children. Pediatrics

1984;73:405–7.

12

Page 33 of 38
Figure 1. Study selection flow diagram according to PRISMA

Page 34 of 38
13 Table 1. Summary of adult cancer patient prophylactic antibiotic lock studies

Control solution Study solution

P-value
Rate of (comparing
Rate of
Study Patient population Infection infection Infections rate of
Lock solution Lock solution infection
per patient per 1000 per patient infection
concentrations concentrations per 1000 per 1000
(%) CVC- (%)
CVC-days CVC-days)
days

Vancomycin-containing solutions

Carratala n=117
et al. V 25 µg/mL +
Neutropenia H 10 units/mL 4/57 (7) 6.37b 0/60 (0) 0 0.05a
(1999) H 10 units/mL
[21] (<500 cells)/
Leukemia=88%
Ethanol-containing solutions

Sanders n=60
Heparinized
et al. 11/30 (36.7) 0.31 70% Ethanol 3/34 (8.8) 0.06 0.008
Leukemia=41% saline
(2008) [8] Multiple
myeloma=35%
n=448
Slobbe et Acute myeloid 16/222 10/226
al. (2010) leukemia plus acute 0.9% NS 1.19 70% Ethanol 0.7 0.19
(7.2)b (4.4)b
[22] lymphocytic
leukemia=58%

14 CVC, central venous catheter; H, heparin; V, vancomycin; CVC, central venous catheter; NS, normal saline.
a
15 P-value used for infection rates per patient.

Page 35 of 38
 b
16 Calculated from published data available in manuscript.

17

Page 36 of 38
18 Table 2. Summary of pediatric cancer patient prophylactic antibiotic lock studies
Control solution Study solution P-value
Study Patient population Lock solution Rate of Rate of Lock solution Rate of Rate of (reported
for rate of
concentrations overall CLABSI concentrations overall infection
CLABSI
CLABSI per per 1000 CLABSI per 1000 per 1000
patient (%) CVC- per patient CVC-days CVC-days)
days (%)
Ethanol-containing solutions
Schoot et al. n=307
(2015) [23] Median age=8.8 years
Solid tumor=51.8% (159) H 100 29/154 16/153
1.46 70% Ethanol 0.77 0.039
Hematological units/mL (19) (10%)
tumour=48.2% (148)
Vancomycin-containing solutions
Schwartz et al. n=45
(1990) [24] Median age=46 months V 25 µg/mL +
H 10 units/mL 6/24 (25) 0.1b 1/21 (4.8) 0.02b 0.035a
Neutropenia on active H 25 units/mL
chemotherapy
Rackoff et al. n=63
(1995) [25] Average age=8 years V 25 µg/mL +
H 100 b b 12/32
Cancer=55% (eight 3/31 (9.7) 0.62 H 100 2.74b NA
units/mL (37.5)b
patients received TPN for units/mL
bowel disorders)
Barriga et al. n=83 patients
V 25 µg/mL + 18/39
(1997) [26] Median age=6 years H 25 units/mL 26/44 (59) 3 2.25 NA
H 25 units/mL (46.1)
Leukemia=72%
Daghistani et n=64
V 25 µg/mL +
al. (1996) [28] Median age=9 years
H 100 H 100
Leukemia=53% (only 61 3/33 (9.1) 0.24 2/28 (7.1) 0.14 NA
units/mL units/mL + A
patients received either
25 µg/mL
arm)
Henrickson et n=126 H 9.73 12/64 (18.8) 0.67 V 25 µg/mL + 1/34 (2.9) 0.09 0.03

Page 37 of 38
 al. (2000) [29] 75% of patients <10 years units/mL H 9.73
Acute leukemia=44% units/mL + C
Solid tumors=40% 2 µg/mL
V 25 µg/ml +
H 9.73 1/28 (3.6) 0.13 0.06
units/mL
Minocycline-containing solutions

Chatzinikolaou n=62
et al. (2003) 80% of patients <14 years
Hematological M 3 mg +
[27] N/A 10/48 (20.8) 2.23 0/14 (0) 0 NA
malignancy=42% Solid EDTA 30 mg
tumor=58%
Taurolidine-containing solutions

Simon et al. n=179

T 13.5 µg/mL
(2008) [30] Median age=9 years
H 100 units 24/90 (27) 4.93 + SC 40 21/89 (24) 3.82 0.35
Leukemia=31%
µg/mL
Solid tumors=29%
Dumichen et n=71 T 13.5 µg/mL
al. (2012) [31] Median age=7 years H 100 units 9/36 (25) 1.3 + SC 40 2/35 (5.7) 0.3 0.03
Leukemia=62% µg/mL +
Handrup et al. n=129 T 13.5 µg/mL
(2013) [32] Median age=6 years + SC 40
H 250 units 26/65 (40) 1.4 7/64 (10.9) 0.4 <0.001
Leukemia/lymphoma=49% µg/mL + H
Solid tumors=51% 100 units/mL
19 CLABSI, central-line-associated bloodstream infection; CVC, central venous catheter; TPN, total parenteral nutrition; H, heparin; V, vancomycin;
20 C, ciprofloxacin; A, amikacin; M, minocycline; EDTA, ethylenediaminetetcetate; T, taurolidine; SC, sodium citrate.
21 NA, P-value not available
a
22 P-value used for infection rates per patient.
b
23 Calculated from published data available in manuscript.

Page 38 of 38