How often should dressings on central venous access devices (CVADs) be changed to reduce catheter-related infection?

Background
A central venous access device (CVAD, also known as a central venous catheter) is a hollow tube that is placed in a large vein with the tip sitting near the heart. CVADs allow medications, fluids and blood products to be given straight into the bloodstream and allow blood samples to be taken for analysis. One of the negative consequences of a CVAD can be an infection of the blood stream which is called catheter-related bloodstream infection or CRBSI which can be serious and even life-threatening. Some CVADs can stay in place for weeks, months or years. Most patients admitted to an intensive care unit will have a CVAD inserted and patients with poor veins or requiring long-term treatment will be offered a CVAD. Dressings are placed over the insertion site of the catheter where it enters the vein, usually in the chest, neck or arm, to protect the surrounding skin. Dressings help prevent infections from starting and they stop the CVAD from moving around. Dressings are changed when they become dirty or they start to fall off. Frequent dressing changes can cause damage to the surrounding skin, so patients may experience pain or skin damage when the dressing is removed. Changing the dressing frequently is also expensive.

We wanted to see if there were any advantages or disadvantages to longer or shorter time intervals between CVAD dressing changes. Some hospitals or healthcare facilities recommend changing dressings every few days, while others keep dressings in place for longer.

Review question
We reviewed the available evidence about the effect of different time intervals between dressing changes for CVADs and whether they had an effect on the risk of CRBSI and other complications. We found five studies that provided information for our review.

Study characteristics
The five studies that were included in the review were published between 1995 and 2009 and involved a total of 2277 participants. Four countries were represented (two studies from France and one each from Italy, Sweden, and the Czech Republic). One study involved children and the remaining four trials included only adults. Four of the studies included cancer patients and one included patients in an intensive care unit.

We classified the time intervals between dressing changes as short (2 - 5 days) in the more frequently changed dressings group and long (5-15 days) in the less frequently changed group. All studies used transparent dressings made of synthetic materials and two studies used gauze (a fabric dressing that does not stick to the skin) secured with tape when skin was damaged. CVAD dressings were monitored on a daily basis in all trials and participants were followed up at least until the CVAD was removed or until discharge. In one study, the manufacturer provided one of the products, but had no influence in the design or how the results were analysed and reported.

Key results
The current evidence leaves us uncertain whether the frequency of dressing changes for CVADs influences risk of CRBSI or death. Of particular interest to patients are problems that may be associated with the dressing themselves, such as pain when they are removed and the skin damage that the dressing may cause. We found no clear evidence that pain, which was assessed daily, was affected by the frequency of dressing changes.

Quality of evidence
The quality of the evidence was very low or low. We downgraded quality because of small and few studies, poor study designs and differences in results between the studies. Better designed studies are still needed to show whether longer interval or shorter intervals between dressing changes are more effective in preventing catheter related infections, mortality, skin damage, dressing removal pain, quality of life and cost.

This plain language summary is up-to-date as of 10 June 2015.

Authors' conclusions: 

The best available evidence is currently inconclusive regarding whether longer intervals between CVAD dressing changes are associated with more or less catheter-related infection, mortality or pain than shorter intervals.

Read the full abstract...
Background: 

People admitted to intensive care units and those with chronic health care problems often require long-term vascular access. Central venous access devices (CVADs) are used for administering intravenous medications and blood sampling. CVADs are covered with a dressing and secured with an adhesive or adhesive tape to protect them from infection and reduce movement. Dressings are changed when they become soiled with blood or start to come away from the skin. Repeated removal and application of dressings can cause damage to the skin. The skin is an important barrier that protects the body against infection. Less frequent dressing changes may reduce skin damage, but it is unclear whether this practice affects the frequency of catheter-related infections.

Objectives: 

To assess the effect of the frequency of CVAD dressing changes on the incidence of catheter-related infections and other outcomes including pain and skin damage.

Search strategy: 

In June 2015 we searched: The Cochrane Wounds Specialised Register; The Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library); Ovid MEDLINE; Ovid MEDLINE (In-Process & Other Non-Indexed Citations); Ovid EMBASE and EBSCO CINAHL. We also searched clinical trials registries for registered trials. There were no restrictions with respect to language, date of publication or study setting.

Selection criteria: 

All randomised controlled trials (RCTs) evaluating the effect of the frequency of CVAD dressing changes on the incidence of catheter-related infections on all patients in any healthcare setting.

Data collection and analysis: 

We used standard Cochrane review methodology. Two review authors independently assessed studies for inclusion, performed risk of bias assessment and data extraction. We undertook meta-analysis where appropriate or otherwise synthesised data descriptively when heterogeneous.

Main results: 

We included five RCTs (2277 participants) that compared different frequencies of CVAD dressing changes. The studies were all conducted in Europe and published between 1995 and 2009. Participants were recruited from the intensive care and cancer care departments of one children's and four adult hospitals. The studies used a variety of transparent dressings and compared a longer interval between dressing changes (5 to15 days; intervention) with a shorter interval between changes (2 to 5 days; control). In each study participants were followed up until the CVAD was removed or until discharge from ICU or hospital.

Confirmed catheter-related bloodstream infection (CRBSI)

One trial randomised 995 people receiving central venous catheters to a longer or shorter interval between dressing changes and measured CRBSI. It is unclear whether there is a difference in the risk of CRBSI between people having long or short intervals between dressing changes (RR 1.42, 95% confidence interval (CI) 0.40 to 4.98) (low quality evidence).

Suspected catheter-related bloodstream infection

Two trials randomised a total of 151 participants to longer or shorter dressing intervals and measured suspected CRBSI. It is unclear whether there is a difference in the risk of suspected CRBSI between people having long or short intervals between dressing changes (RR 0.70, 95% CI 0.23 to 2.10) (low quality evidence).

All cause mortality

Three trials randomised a total of 896 participants to longer or shorter dressing intervals and measured all cause mortality. It is unclear whether there is a difference in the risk of death from any cause between people having long or short intervals between dressing changes (RR 1.06, 95% CI 0.90 to 1.25) (low quality evidence).

Catheter-site infection

Two trials randomised a total of 371 participants to longer or shorter dressing intervals and measured catheter-site infection. It is unclear whether there is a difference in risk of catheter-site infection between people having long or short intervals between dressing changes (RR 1.07, 95% CI 0.71 to 1.63) (low quality evidence).

Skin damage

One small trial (112 children) and three trials (1475 adults) measured skin damage. There was very low quality evidence for the effect of long intervals between dressing changes on skin damage compared with short intervals (children: RR of scoring ≥ 2 on the skin damage scale 0.33, 95% CI 0.16 to 0.68; data for adults not pooled).

Pain

Two studies involving 193 participants measured pain. It is unclear if there is a difference between long and short interval dressing changes on pain during dressing removal (RR 0.80, 95% CI 0.46 to 1.38) (low quality evidence).