CRD summary

The authors concluded that there was some evidence that interventions other than antimicrobial catheters (such as staff education, multifaceted programmes and performance feedback) could reduce catheter-related bloodstream infections in adult intensive care unit patients. These specified strategies appeared supported by limited evidence from a small number of generally flawed observational studies. A more cautious conclusion may have been more appropriate.

Authors' objectives

To evaluate strategies for reducing catheter-related bloodstream infections (CRBSI) in intensive care unit patients other than antimicrobial coated catheters.

Searching

The following databases were searched for studies published between 1985 and February 2007: MEDLINE via PubMed; CINAHL; Current Contents; Current Contents Connect; Australia Medical Index; Biological Abstracts; EMBASE; Science Citation Index; National Library of Medicine; Dissertation Abstracts; DARE; The Cochrane Library; Health Services Technology; National Clearing House; Centre for Disease Control guidelines and reports; Bandolier; and Clinical Evidence. Search terms were reported. Reference lists of relevant guidelines, reviews and included studies were screened. Only studies with an English-language abstract and that had been published in full were included.

Study selection

Randomised controlled trials (RCTs) and observational studies that evaluated short-term (<21 days) non-tunnelled catheters in adult intensive care unit patients and reported the incidence of CRBSI were eligible for inclusion. Studies were only included if they defined CRBSI using criteria specified in the review; the review also assessed catheter colonisation defined as specified in the review. Studies that evaluated totally implanted catheters, peripherally inserted central venous catheters or haemodialysis catheters were excluded.

The review evaluated eight different interventions: site of central venous catheter insertion; choice of skin disinfectant; catheter replacement at a new site versus exchange over a guidewire; connectors and hubs; attachable cuffs; number of lumens on central venous catheter; educational programmes for healthcare professionals; and combined interventions. Studies included medical, medical/surgical, cardiac, surgical, cardiac surgical and neurosurgical patients. Where reported, the mean age ranged from 42 to 71 years and the percentage of males ranged from 46% to 81%.

Two reviewers independently selected studies.

Assessment of study quality

Study validity was assessed using criteria derived from two specified validated checklists. Studies were classified into the following three groups: 1) well-designed study with generalisable findings; 2) acceptable experimental study with findings that may be generalised; and 3) flawed study rejected from review. RCTs were additionally assessed for reporting of randomisation, allocation concealment, blinding, intention-to-treat analysis and sample size calculation.

The authors did not state how the validity assessment was performed.

Data extraction

Two blinded reviewers independently extracted data. Authors were contacted if required. For each study, risk ratios with 95% confidence intervals (CI) were calculated.

Methods of synthesis

Studies were grouped by type of intervention and generally combined in a narrative synthesis. Data from similar studies were pooled using the random-effects DerSimonian and Laird method. Heterogeneity was assessed using the Mantel-Haenszel test statistic. There were too few studies to perform sensitivity and subgroup analyses.

Results of the review

Additional results from RCTs and results from observational studies were reported in the review.

Twenty-three studies were included (n was at least 5,885 patients or central venous catheters): 13 RCTs (n=1,945), one non-randomised trial (n=154) and nine observational studies (n>3,786).

Study quality: One RCT reported allocation concealment, four reported blinding of analysis personnel, three used intention-to-treat analysis and six reported a sample size calculation.

Site of central venous catheter insertion (two studies): One RCT reported significantly higher rates of colonised catheters inserted through the femoral compared to the subclavian vein (p<0.001). One observational study reported no significant difference between internal jugular and axillary vein insertion.

Skin disinfectant (three studies): One RCT reported a significant reduction in colonisation associated with 2% aqueous chlorhexidine compared to 10% povidone iodine ( p=0.01), but no significant difference between 70% alcohol and 2% chlorhexidine. One RCT reported no significant difference between 10% povidone iodine and 0.5% chlorhexidine. One observational study reported significantly lower colonisation rates associated with alcoholic povidone iodine compared with aqueous povidone iodine (p<0.001).

Attachable cuffs (two studies): Triple-lumen catheters with attachable cuff (VitaCuff) were associated with a significant reduction in colonisation and a reduction in CRBSI compared to catheters with no cuff in one RCT (p=0.02), but no significant difference between interventions in one non-randomised study.

Studies reported no difference for catheter replacement at a new site versus exchange over a guidewire (one RCT), between different types of hubs and connectors (four RCTs) or between catheters with different numbers of lumens (two RCTs and two observational studies).

Educational programmes for healthcare professionals (one study): One pre-post study reported that an educational intervention was associated with a significant reduction in CRBSI from baseline,

Combined interventions (six studies): There was no significant difference between various combinations of silver-impregnated implantable collagen cuffs versus no cuff and a catheter removal policy at three or seven days (one RCT), or between different modes of catheter replacement and replacement schedules (one RCT). One observational study reported a significant reduction in catheter colonisation from baseline in a phased intervention (p=0.01), but no significant reduction in CRBSI. Two of three phased pre-post test studies that evaluated combinations of education, training, catheter policies and feedback reported significant reductions in CBRSI from baseline (p=0.0001 and p<0.001); one study found no significant reduction post-intervention.

Cost information

The authors stated that these interventions may be less costly to implement than widespread use of antimicrobial catheters, but further analysis of their cost-effectiveness was required; no cost data were reported in the review

Authors' conclusions

There was some evidence that the risk of CRBSI can be reduced by interventions other than antimicrobial catheters such as staff education, multifaceted infection control programmes and performance feedback.

CRD commentary

The review question was clearly stated and inclusion criteria were specified for participants, interventions, study design and outcomes. Many relevant sources were searched, but only studies published in English were eligible and this raised the potential for publication and language biases. Study validity was assessed and results for RCTs were reported. Appropriate methods were used to minimise reviewer error and bias during study selection and data extraction, but it was unclear whether similar methods were used for the validity assessment. In view of the diversity among studies, a narrative synthesis with studies grouped by intervention type was appropriate. Evidence from RCTs was indicated, but the quality of RCTs was not taken into account when summarising the data. Evidence was based on a small number of studies that appeared generally to be of limited quality. Results for some interventions were mixed. The authors’ conclusions on effective strategies appeared to be supported by limited evidence from a small number of generally flawed observational studies and a more cautious conclusion may have been more appropriate.

Implications of the review for practice and research

Practice: The authors stated that the recommended strategies were low cost, low risk and constituted good nursing practice, but they did not appear to make clear recommendations

Research: The authors stated that future studies should be adequately powered and be carefully designed to minimise potential sources of bias. There was a need for cost-effectiveness of all strategies designed to reduce CRBSI to be evaluated.

Funding

National Health and Medical Research Council of Australia.

Bibliographic details

Ramritu P, Halton K, Cook D, Whitby M, Graves N. Catheter-related bloodstream infections in intensive care units: a systematic review with meta-analysis. Journal of Advanced Nursing 2008; 62(1): 3-21. [PubMed: 18352960]

Original Paper URL

http://onlinelibrary.wiley.com/journal/119403780/abstract

Indexing Status

Subject indexing assigned by NLM

MeSH

Adult; Bacterial Infections /prevention & control; Catheterization, Central Venous /adverse effects /statistics & numerical data; Catheters, Indwelling /adverse effects /microbiology /statistics & numerical data; Clinical Competence /standards; Cost-Benefit Analysis; Cross Infection /prevention & control; Equipment Contamination; Female; Humans; Incidence; Infection Control /standards; Intensive Care /standards; Male; Middle Aged; Risk Factors; Subclavian Vein

AccessionNumber

12008104140

Database entry date

23/09/2009

Record Status

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.