Antimicrobial Catheter Lock Solutions

STUDY PURPOSE: To investigate antibiotics or other lock treatments in comparison to a control intervention for the treatment of central venous catheter (CVC)-related infections in children with cancer

TYPE OF STUDY: Systematic review

DATABASES USED: CENTRAL (2011), MELINE/PubMed (1945–August 2011), and EMBASE/Ovid (1980–August 2011); reference lists from relevant articles and conference proceedings (SIOP, 2006–2010; American Society of Clinical Oncology, 2006–2010; Multinational Association of Supportive Care in Cancer, 2006–2011; American Society of Hematology, 2006–2010; and International Society on Thrombosis and Haemostasis, 2006–2011); and scanned ISRCTN for ongoing trials

KEYWORDS: See appendices attached.

INCLUSION CRITERIA: Randomized controlled trials (RCTs) and controlled clinical trials (CCTs) comparing one lock treatment with another, or with systemic antibiotics alone, to treat CVC-related infections in children with cancer; cohort studies also were included for adverse events.

EXCLUSION CRITERIA: Adults included in the analyses, studies focused on infection prevention, use of coated catheters, observational studies, no control comparisons, case report studies, and review studies

TOTAL REFERENCES RETRIEVED = 508

EVALUATION METHOD AND COMMENTS ON LITERATURE USED: Two review authors searched the databases and reviewed articles for inclusion criteria. Bias assessment was conducted, and if no agreement was reached, a third party was included in the assessment. Pooled analysis was conducted using Review Manager 5 when studies met methodologic quality and were comparable; otherwise, descriptive summaries were conducted. Comparisons included the creation of forest plots. The comprehensive search of multiple databases, meeting proceedings, and ongoing trials to elicit articles for the evaluation of RCTs, CCTs, and cohort studies was appropriate for this evaluation.

• FINAL NUMBER STUDIES INCLUDED = 12 (not all included in all of the analyses)
• SAMPLE RANGE ACROSS STUDIES = 23–68
• TOTAL PATIENTS INCLUDED IN REVIEW = 132
• KEY SAMPLE CHARACTERISTICS: Children with cancer (aged 0–18 years) who incurred a CVC-related infection

• PHASE OF CARE: Active antitumor treatment     
• APPLICATIONS: Pediatrics

No significant differences were found between the use of ethanol or urokinase lock treatments with systemic antibiotics and systemic antibiotics alone for the outcomes of number cured, number of recurrent CVC-related infections, days until first negative blood culture, number of early removal of CVCs, intensive care unit admissions, or sepsis.

No benefit of urokinase or ethanol lock in addition to systemic antibiotics was found.

• Small sample sizes
• Short follow-up time
• Some CVC malfunctioning was reported.
• No adverse events were reported in the cohort studies.
• No RCTs or CCTs that used antibiotic lock treatment alone were found in the literature.

No added benefit was found in the use of urokinase or ethanol lock treatment in addition to systemic antibiotics for the treatment of CVC-related infections in children undergoing treatment for cancer.

The purpose of the article was to summarize the evidence on the effectiveness of antibiotic-based catheter-lock solutions as compared to heparin-lock solutions to prevent catheter-related blood stream infections (CRBSI) in all patients with long-term intermittent use of central venous catheters (CVCs).

Medline and Cochrane Central Register of Controlled Trials (CENTRAL) databases were searched.

Trials that were included were planned as randomized, controlled trials, quasi-randomized trials or systematic reviews/meta-analyses of randomized or quasi-randomized trials, or published as an article. Trials were also included if the effects of one or more preventative antibiotic-based lock solutions were studied in patients with CVCs for intermittent use, and the presentation of sufficient data for calculating risks of CRBSI in the treatment and control group was examined.

No exclusion criteria were specified.
 

Twenty-three total references were retrieved.

Two reviewers independently assessed trial quality using the following components: concealment of allocation, blinding during treatment and at outcome assessment, description of drop-outs, and analysis. Only trial data related to the topic of the review were considered.

The included trials were flawed due to shortcomings that could have introduced bias, as only 2 of 16 trails clearly prevent performance bias and, in eight of those trials, methods of blinding were unclear. Nine  trials had unclear allocation concealment and only one trial performed analysis by intention to treat. Baseline comparability of groups did not differ. Design and methodology of included studies were sufficient to analyze and pool data.
 

• Sixteen studies were included in the final review.
• 1,383 catheter days were included from 532 patients with cancer (261 were children).
• Catheter days across all included studies ranged from 14–80.
• Key sample characteristics were not specified.

• The phase of care was active treatment
• Applications were for late effects, survivorship, and pediatrics

Overall, it could not be determined which antibiotic-based lock solution is most effective in reducing CRBSI. Only two small trials compared different antibiotics head-to-head. In hemodialysis patients, there was a significant benefit in favor of the antibiotic-based solutions in patients with cuffed or tunneled catheters.  In pediatric oncology patients, there was a small but statistically significant benefit of the antibiotic-based lock solutions in the prevention of BSI (not CRBSI). There was an overlap of 42 elderly patients between two trials. Vancomycin-containing lock or flush solutions are effective in reducing the risk of BSI in patients with cancer. One trial also demonstrated a significant reduction of gram-positive CRBSI, using vancomycin flush solutions in pediatric patients with cancer. 

• The studies should be interpreted with caution as all had a small number of patients.
• There was a difference in solutions that were flushed into the bloodstream and lock solutions, and these were not differentiated in this review.  

Although some results seemed promising, these should be interpreted with care, especially in patients with cancer. There was no differentiation made between BSI and CRBSI, which could be complicated in interpreting the results. There also is a small sampling in regard to patients with cancer, as there were only six oncology trials included in the review. There are no real indications that the use of antibiotic-based lock solutions could prevent catheter-related infections, and it is not possible to determine which antibiotic-based lock solution is most effective. There is the risk of bacterial antibiotic resistance when using broad spectrum antibiotics for locking solutions. This should be weighed against the benefit of locking or flushing solutions.

STUDY PURPOSE: To determine the effectiveness of administering antibiotics prior to long-term central venous catheter (CVC) insertion, or flushing to prevent catheter-related infections

TYPE OF STUDY: Meta-analysis and systematic review

• FINAL NUMBER STUDIES INCLUDED = 11
• TOTAL PATIENTS INCLUDED IN REVIEW = 828
• SAMPLE RANGE ACROSS STUDIES: 27–108 patients
• KEY SAMPLE CHARACTERISTICS: Patients with varied tumor types

PHASE OF CARE: Not specified or not applicable
 
APPLICATIONS: Pediatrics

Five studies evaluated prophylactic antibiotic administration prior to CVC insertion, and six studies evaluated flush or catheter lock solutions. Pooled analysis of evidence comparing antibiotic and heparin solution to heparin only solution showed that the combination of antibiotic was associated with less catheter-related sepsis than the heparin only solution (468 participants, relative risk [RR] = 0.47, 95% confidence interval [CI] [0.28, 0.8], p = 0.0051). The majority of these instances were in children, and low heterogeneity existed among these studies. Five of the six studies used vancomycin in the flush mixture. No difference existed in the risk of catheter-related infections between those who did and did not receive prophylactic antibiotics prior to CVC insertion.

This review did not demonstrate a beneficial effect of prophylactic antibiotics prior to long-term CVCD insertion. The findings suggest that the use of antimicrobial catheter flush or lock solutions may be helpful for infection reduction from gram-positive organisms; however, the majority of studies used vancomycin, which has been not recommended for standard use.

• Low sample sizes
• Lack of specificity and consistency in outcomes

The use of catheter flush and lock solutions that contain a combination of antibiotics and heparin may help prevent or reduce catheter-related infection in patients with long-term tunneled central venous catheters.

The purpose of the study was to evaluate the efficacy of using heparin versus M-EDA for locking central venous catheters and preventing infection.

Prospective blood cultures were obtained at the beginning of the study and at each chemotherapy session at weekly or monthly intervals according to each chemotherapy protocol. Aseptic technique was used to collect 10 ml of blood for each culture. Cultures obtained from the catheters also were obtained with catheter removal. Primary outcome was positive blood culture or clinical evidence of bacteremia or sepsis associated with the catheter, regardless or whether blood cultures were positive or negative. The catheter-locking solution had the same volume as each catheter’s priming solution. It was slowly introduced (in at least 10 seconds) after each chemotherapy session, and remained in the catheter lumen until the next session.

• The sample consisted of 50 patients, 26 in the heparin arm (group 1) and 24 in the M-EDTA arm (group 2).
• Age range was not given, but patients were classified as pediatric.
• All patients with blood cancers and solid tumors (group 1 contained 16% blood cancers and 10% solid tumors; group 2 contained 16% blood cancers and 8% solid tumors) were included.
• All patients were undergoing chemotherapy, had an implanted port-a-cath CVC, and were clinically stable.
   

Single-site inpatient/outpatient facility in Sao Paulo, Brazil
 

• The phase of care was active treatment
• Application was for pediatrics and CVCs
   

Prospective, randomized study

Blood cultures

A total of 762 serial prospective blood cultures were obtained, 387 from group 1 and 375 from group 2. In group 1, 19 blood cultures were positive and infection incidence was 73.1% (19 of 26 ports); in group 2, five blood cultures were positive and the incidence rate was 20.8% (5 of 24 ports). This difference was significant (p = 0.0001). The colonization of catheters was 5.7 times greater in group 1 than group 2. The mean time free of catheter infection in group 1 was 4.72 months, significantly shorter than in group 2, where it was 9.69 months (p = 0.002). The chances of hospitalization were two times greater for children in the heparin group. There were no side effects observed from either treatment.

The results of this study suggest a significant benefit for children using M-EDTA as a CVC lock as opposed to straight heparin for prevention of catheter-associated infections. The EDTA is a strong cation with chelating properties that destroys the buildup within the lumen. Minocycline is a great broad spectrum antibiotic that is never used for anything but orally for acne.

• Small sample size (less than 100 participants), and too few completed study. 
• No blinding with associated risk of bias.
• It is unclear if differences in hospitalizations between the two groups were related to infection, or if the infection differences seen may have been in part a result of more hospitalizations.
• Specific ingredients and proportions in the solution used are not provided. 
• No subgroup analysis was conducted for hematologic and solid tumors, where infection risk may be very different.  
   

The nurse would not be able to use this without a physician's order, unless it was SOP for that facility. If it were SOP, then nurses could use the solution to flush and lock the CVCs.

This approach shows great promise and would be simple to implement in a clinical setting. Further research is warranted to provide additional support for its efficacy.

The purpose of the study was to compare the effect of catheter locking with taurolidine versus heparin in biofilm formation in central venous catheters.

In the standard arm, catheters were locked with 250 IU heparin in 2.5 ml normal saline while in the experimental arm they were locked with taurolidine 2.5 ml in a sodium citrate and heparin solution. All catheters were either tunneled or totally implanted devices and chosen at the physician’s discretion. Biocclusive dressings were changed every three days and the skin was cleansed with chlorhexidine with dressing changes.

• The sample contained 48 participants, aged 0–19 years.
• All participants were males
• Patients had any oncological disease requiring a tunneled CVC

A single-site inpatient setting in Denmark

• The phase of care was active antitumor treatment
• Application was for pediatrics

Prospective, randomized, controlled, open-labeled study

• After removal, catheters were examined by standardized scanning electron microscopy to assess quantitative biofilm formation. Quantitative and semi-quantitative cultures were also performed using scanning electron microscopy (SEM).
• Catheter-related bloodstream infection (CLABSI) was defined as microbe growth from a blood sample drawn for the CVC at least two hours before microbial growth was detected in a peripheral vein blood sample.

There was no significant difference in the formation of biofilm between the two groups (p = 0.13). A reduction in catheter-related blood stream infections (CRBSIs) was demonstrated in the taurolidine arm (p = 0.03). CVCs locked with heparin were removed after a median of 246 days (range = 40–1,081) and after a median of 301 days (range = 51–590) in those with the experimental lock solution.

The trial confirmed that use of taurolidine as catheter-lock compared with heparin reduced the rate of CRBSIs. This reduction was not related to a reduction in the biofilm formation.

• Small sample size (less than 100)
• Risk of bias (no blinding)
• Risk of bias(sample characteristics)
• Measurement validity/reliability questionable
• All participants in this study were male.
• Also, the number of hematological malignancy were unevenly distributed with a large portion of these patients in the experimental arm. This may have contributed to a higher risk of infection.
• Dehydration of specimens for SEM caused the specimens to become fragile, potentially affecting the accuracy of the study.

No difference in CVC survival was noted, requiring that they will be changed at the same rate as before. Findings suggest that taurolidine used as a catheter lock was associated with lower incidence of CRBSI.

To determine if a taurolidine catheter lock can reduce catheter-related bloodstream infection (CRBSI) in children with tunneled central venous catheters (CVCs)

Patients were randomized to receive either locks with 250 IE heparin in 2.5 ml normal saline or with 2.5 ml taurolidine 1.35%/sodium citrate 4%/heparin 100 IE/ml. Catheters were flushed once weekly. Catheter insertion was done according to standards in all patients, and bio-occlusive dressings were changed weekly after the skin was cleansed with chlorhexidine every three days. Tunneled lines and total implantable devices were included.

• N = 112 patients, 129 catheters  
• MEDIAN AGE = 6 years
• AGE RANGE = 0–19 years
• MALES: 61%, FEMALES: 39%
• KEY DISEASE CHARACTERISTICS: Multiple tumor types; the most common was leukemia or lymphoma.
• OTHER KEY SAMPLE CHARACTERISTICS: 113 of the 129 catheters were totally implantable devices.

• SITE: Single site   
• SETTING TYPE: Not specified   
• LOCATION: Denmark

• PHASE OF CARE: Multiple phases of care
• APPLICATIONS: Pediatrics

• Randomized, two-group, open-label design

• Tunnel infection was defined as inflammation extending more than 2 cm along the tract of the CVC.
• CRBSI was defined as growth of microbes from a sample drawn from the CVC at least two hours before microbial growth was detected in a peripheral blood sample, or cultures in which a pathogen was found in the central sample but not in the peripheral sample.
• Probable CRBSI was determined if a recognized pathogen was cultured from at least one blood sample, or a common skin contaminant was cultured from two or more samples obtained at separate occasions.

There were 33 episodes of CRBSI. The rate of total bloodstream infections per CVC days was seen in those with taurolidine locks (1.2 per 1,000 CVC days) compared to those with heparin locks (2.5 per 1,000 CVC days) (IRR = 0.49. p =.004). The rate of CRBSI in the experimental group was 0.4/1,000 CVC days compared to 1.4/1,000 CVC days (IRR = 0.26, p = .001). CVC survival was similar in both groups, with a median of 256 days in the heparin group and 300 days in the taurolidine group. Power analysis showed that the sample size was sufficient to detect a relative risk of 0.25 with the intervention.

Use of taurolidine citrate catheter locks was effective in preventing CRBSI in pediatric patients with long-term CVCs. The majority of these were totally implantable devices.

• Risk of bias (no blinding)
• Risk of bias(sample characteristics)
• Other limitations/explanation: Open-label design; the majority of patients had totally implantable devices, so it is not clear if findings would apply similarly to only tunneled catheters.

CRBSI is a major concern for patients with cancer who are immunocompromised. Results of this study provide an intervention that appears to prevent CVC-related infections with long term CVCs. Because the majority of catheters in this study were totally implantable devices, it is not clear if this will apply to other long- or short-term CVCs, but further study in these areas is warranted.

To determine if 70% ethanol catheter locks can prevent central line–associated bloodstream infections (CLABSI) in pediatric patients with cancer

Patients were randomised to receive either 70% ethanol or 100 IU/ml heparin locks. Locks were used only when needed, no more than weekly if not accessed, and at least every six weeks. Catheter insertion and care were done according to international guidelines. Patients were stratified according to type of catheter, port-a-cath or broviac type, and type of disease. Two hours after the lock solution was used, the catheter was flushed with saline and the CVC was closed with heparin.

• N = 237
• MEAN AGE = 8.8 years
• AGE RANGE = 1–18 years
• MALES: 57%, FEMALES: 43%
• OTHER KEY SAMPLE CHARACTERISTICS: 94% had single-lumen catheters. The majoirty of them (51%) were placed in the right jugular vein.

• SITE: Multi-site  
• SETTING TYPE: Not specified  
• LOCATION: Netherlands

• APPLICATIONS: Pediatrics

• Single-blind randomized, controlled trial

• Central line–associated bloodstream infection (CLABSI) defined as pathogen cultured from at least one blood culture and fever, chills, or hypotension and a common skin organism from at least two culture drawn on separate locations

The total number of catheter days were 20,916 in the ethanol group and 19,915 in the heparin group. In the ethanol group, 10% of patients developed a CLABSI, compared to 19% in the heparin group (hazard ratio [HR] = 0.52, p = 0.05). CLABSI incidence per catheter days was 0.77/1000 in the ethanol group and 1.46/1000 in the heparin group (p = 0.039); however, when corrected for type of catheter, there was no significant difference between groups. Patients with internalized catheters had significantly lower risk of CLABSI (HR = 0.22, p < 0.001). Survival curve analysis at 100 days showed lower CLABSI rate in the ethanol group (p = 0.038). There was no difference between groups in the number of positive blood cultures. There was no difference in hospital days related to infection. Patients in the ethanol group reported nausea, taste alteration, and dizziness during the time of infusion or lock flushing. A larger proportion of those in the ethanol group withdrew from the study (p = 0.031).

Although authors concluded that ethanol locks reduced CLABSI rate, there was no difference when analyzed according to type of catheter, whether indwelling or not. It appears that the type of catheter may be more important for CLABSI prevention.

• Key sample group differences that could influence results
• Measurement/methods not well described
• Subject withdrawals ≥ 10%  
• No information about proportion of study groups that had indwelling catheters
• There were high catheter days in the ethanol group.
• Timing of study measures was not provided.

Findings of this study are not conclusive regarding the efficacy of two-hour ethanol CVC locks, though results are somewhat promising. Further research is needed with clear differentiation of indwelling catheter versus broviac catheters in analysis. This study used a dwell time of two hours for the ethanol lock; the optimum dwell time is unclear.

To provide guidance for clinical practices for the prevention and treatment of infection in patients with cancer.

This resource is a consensus-based guideline.

Patients were undergoing the active antitumor treatment phase of care.

The guideline

• Recommends the consideration of general antibacterial prophylaxis in patients at intermediate and high risk for infection, consideration of antifungal prophylaxis during neutropenia and for anticipated mucosits, and antiviral prophylaxis for intermediate- and high-risk patients.
• Provides suggestions for specific agents for prophylaxis and treatment in various clinical scenarios.
• Outlines treatment and diagnostic/assessment approaches for neutropenic fever and specific clinical presentations.
• Notes that chlorhexidine and sliver sulfadiazine-coated short-term central catheters have been shown to decrease the incidence of catheter colonization and bloodstream infections, but not in patients with hematologic malignancies requiring indwelling catheters for approximately 20 days.
• Notes that vaccination recommendations for transplantation recipients and their household members should be performed.
• Recommends the pneumococcal vaccine in asplenic patients.

The National Comprehensive Cancer Network (NCCN) does not currently endorse the use of a vancomycin lock solution for long-term vascular access devices due to concerns about the emergence of bacterial resistance if widely used. Influenza vaccination with a vaccine that does not use live attenuated organisms can be safely given, and the guideline recommends administration at least two weeks before receiving cytotoxic therapy.

This study lacked high-quality evidence, with most recommendations being based on consensus.

This guideline provided comprehensive references to assess patient risk of infection and expert recommendations regarding interventions aimed at the prevention and treatment of infection in patients with cancer. The guideline does not discuss long-term survivorship issues in this area.

To provide evidence-based recommendations for the prevention of intravascular catheter-related infections for healthcare personnel who insert and care for intravascular catheters and for those responsible for surveillance and infection control in hospital, outpatient, and home health settings. Patients addressed in the guidelines were adult and pediatric patients with intravascular catheters.

The resource is comprised of evidence-based guidelines. For the development process, evidence was categorized as category 1A to category 2 based on strength of recommendation and support from clinical or epidemiological studies.

• Patients were undergoing multiple phases of care.
• The study has clinical applicability for pediatrics. 

The results were not summarized.

The guidelines provided extensive recommendations regarding the education and training of staff; selection of catheters and sites, including avoidance of the femoral vein for central venous access and use of the central venous catheter with the minimum number of ports needed; hand hygiene; use of maximal sterile barrier precautions for insertion; skin preparation with alcohol, iodine, or chlorhexidine; use of standard catheter site dressing regimens; specific aspects of care for umbilical and dialysis catheters; and use of piggybacks, stopcocks, and catheter flushing. Guideline recommendations include a bundling of multiple recommendations: antimicrobial-impregnated catheters and cuffs are recommended in patients with long-term use if the organizational central line-associated blood stream infection (CLABSI) rate is not decreasing despite the implementation of comprehensive strategies for improvement. Lower-level (category II) recommendations include the use of prophylactic antimicrobial lock solutions in patients with long-term catheters who have a history of CLABSIs despite optimal aseptic technique as well as daily cleansing of patients in the intensive care unit with 2% chlorhexidine-impregnated washcloths. Recommends changing IV administration sets no more often than every 96 hours unless used for blood products. For needleless components, recommends changing according to administration set timing as above, and states no benefit to changing more than every 72 hours.

Provides extensive recommendations for management of all types of intravenous catheters and system components. Nurses should refer to the full set of guidelines for all specific aspects of care.