Prophylactic antibiotics for preventing gram-positive infections associated with long-term central venous catheters in adults and children receiving treatment for cancer

The issue
People undergoing anti-cancer treatment (chemotherapy) often have a catheter (tube) inserted into a large vein, through which their chemotherapy is given. This is known as a central venous catheter (CVC). As chemotherapy is usually administered over several months to years, long-term CVCs are used. Despite sterile insertion and care afterwards, these CVCs often become infected by a type of bacteria called gram-positive bacteria. Giving antibiotics as a preventive (prophylactic antibiotics) before the catheter is inserted, or using the antibiotics as a lock or flush solution, or both, may prevent these infections. A lock/flush solution is instilled in the catheter when it is not in use. This solution often contains heparin, saline, or both, to prevent a clot forming in the catheter. 

The aim of the review
The aim of this review was to determine whether giving antibiotics before catheter insertion or as a lock/flush solution can prevent gram-positive CVC-related infections.

What are the main findings?
We searched databases to identify randomised controlled trials (RCTs - clinical studies where people are randomly put into one of two or more treatment groups). We identified 12 relevant studies, which included a total of 1244 adults and children with cancer.

We included six studies (756 people) that compared giving antibiotics before the insertion of the CVC against not giving antibiotics. We found that giving an antibiotic before CVC insertion did not prevent infections. The studies authors' did not describe side-effects. 

We included six studies (488 people) that compared an antibiotic solution with a heparin-only solution to flush/lock a CVC. We found that an antibiotic lock/flush solution reduced the number of CVC-related infections. People in one study reported an unpleasant taste after flushing, and one study found there were fewer CVC obstructions (e.g. a clot) in people who received the combined antibiotic and anticoagulant solution. 

Certainty of the evidence
We have moderate confidence in the evidence for giving antibiotics prior to CVC insertion and antibiotic lock/flush solutions since the studies were done in different people, used different types of catheters and antibiotics, and measured their results in different ways. 

What are the conclusions?
We did not observe any benefit of giving antibiotics before the insertion of long-term CVCs to prevent gram-positive CVC-related infections. Flushing/locking long-term CVCs with an antibiotic appeared to reduce gram-positive CVC-related infections in people with cancer. It must be noted that the use of an antibiotic flush/lock solution may increase microbial antibiotic resistance, therefore it should be reserved for high-risk people or if baseline CVC-related infection rates are high. 

This review updates our previous review. The evidence is up to date to 19 November 2020.

Authors' conclusions: 

Since the last version of this review, we included one additional study. There was no observed benefit of administering antibiotics before the insertion of long-term CVCs to prevent gram-positive CVC-related infections. Flushing or locking long-term CVCs with an antibiotic solution likely reduces gram-positive CVC-related infections experienced in people at risk of neutropenia through chemotherapy or disease. However, a limitation of this review is heterogeneity between the studies for both outcomes. Insufficient data were available to evaluate if the conclusions apply equally for different CVC types and for adults versus children. It must be noted that the use of an antibiotic flush/lock solution may increase microbial antibiotic resistance, therefore it should be reserved for high-risk people or if the baseline CVC-related infection rates are high. Further research is needed to identify high-risk groups most likely to benefit from these antibiotic flush/lock solutions. 

Read the full abstract...
Background: 

This is an updated version of a Cochrane Review last published in 2013. Long-term central venous catheters (CVCs), including tunnelled CVCs (TCVCs) and totally implanted devices or ports (TIDs), are increasingly used when treating people with cancer. Despite international guidelines on sterile insertion and appropriate CVC maintenance and use, infections remain a common complication. These infections are mainly caused by gram-positive bacteria. Antimicrobial prevention strategies aimed at these micro-organisms could potentially decrease the majority of CVC-related infections. The aim of this review was to evaluate the efficacy of prophylactic antibiotics for the prevention of gram-positive infections in people with cancer who have long-term CVCs.

Objectives: 

To assess the effects of administering antibiotics prior to the insertion of long-term CVCs or as a flush/lock solution, or both during long-term CVC access to prevent gram-positive CVC-related infections in adults and children receiving treatment for cancer.

Search strategy: 

The search for this updated review was conducted on 19 November 2020. We searched the Cochrane Central Register of Controlled Trials (CENTRAL) in the Cochrane Library, MEDLINE via Ovid and Embase via Ovid. We searched ClinicalTrials.gov and the World Health Organization International Clinical Trials Registry Platform portal for additional articles.

Selection criteria: 

We included randomised controlled trials (RCTs) that compared either the administration of prophylactic antibiotics prior to long-term CVC insertion versus no administration of antibiotics, or the use of an antibiotic versus a non-antibiotic flush/lock solution in long-term CVCs, in adults and children receiving treatment for cancer.

Data collection and analysis: 

We used standard methodological procedures expected by Cochrane. Two authors independently selected studies, classified them and extracted data onto a predesigned data collection form. The outcomes of interest were gram-positive catheter-related infection events and total number of CVCs and CVC days. We pooled the data using a random-effects model for meta-analyses. We used the GRADE approach to assess the certainty of the evidence. 

Main results: 

For this update, we identified 310 potentially relevant studies and screened them for eligibility. We included one additional RCT with 404 participants. The original review included 11 RCTs with a total of 840 people with cancer (adults and children). In total this review included 12 RCTs with 1244 participants.

Antibiotics prior to insertion of the CVC
Six trials compared the use of antibiotics (vancomycin, teicoplanin, ceftazidime or cefazolin) versus no antibiotics given before the insertion of a long-term CVC. One study did not observe any CVC-related infection events in either group was not included in the quantitative analysis as it was not possible to calculate a risk ratio. Administering an antibiotic prior to insertion of the CVC may not reduce gram-positive CVC-related infections (pooled risk ratio 0.67, confidence interval (CI) 95% 0.32 to 1.43; control versus intervention group risk 10.4% versus 7.3% of the participants; 5 studies, 648 participants; moderate-certainty evidence). We sought adverse event data, but these were not described by the authors. The overall risk of bias was deemed low.  

Antibiotics as a flushing or locking solution
Six trials compared a combined antibiotic (vancomycin, amikacin or taurolidine) and heparin solution with a heparin-only solution for flushing or locking the long-term CVC after use. One study did not observe any CRS events and was not include this study in the quantitative analysis as it was not possible to calculate a risk ratio. Flushing and locking long-term CVCs with a combined antibiotic and heparin solution likely reduced the risk of gram-positive CVC-related infections compared to a heparin-only solution (pooled rate ratio 0.47, CI 95% 0.26 to 0.85; control versus intervention group rate ratio 0.66 versus 0.27 per 1000 CVC-days; 5 studies, 443 participants; moderate-certainty evidence). One trial reported a higher incidence of occlusions and participants in one trial reported an unpleasant taste after flushing associated with a combined antibiotic and heparin solution. The overall risk of bias was deemed low.