Central venous catheter-related infection: a cohort study evaluating dedicated central venous catheter packs

Introduction

Central venous catheter (CVC)-related bloodstream infections (CRBSI) are the third most common healthcare-associated infection (HAI) in ICUs, associated with significant morbidity, mortality, increased length of stay and costs [1, 2]. Several care bundle studies have suggested that utilising various strategies together (such as training, regular lines monitoring and using dedicated line insertion trolleys) can have a positive impact on CRBSI rates [3–6]. However, the impact solely attributable to the provision of a dedicated, stand-alone CVC insertion pack has not been evaluated. We therefore investigated the impact of a new EPIC2 compliant CVC pack, introduced in Good Hope Hospital in 2007, on CVC tip colonisation rates.

Methods

Data were collected prospectively between June 2007 and December 2008. Patients were divided into two cohorts: patients whose CVCs were inserted using the CVC packs (B), and those receiving CVCs prior to the introduction of the packs (nB). Data were collated from questionnaires as well as patients' notes on: patient's age and sex; type and site of CVC inserted; location at the time of insertion; the grade of practitioner; and the duration CVCs remained in situ. Using the hospital's patient information system and patient notes, data on CVC tip cultures were obtained. Data are presented as percentages and analysed using multivariate analysis.

Results

Complete data were obtained for 347 patients: 246 patients in group nB, 101 patients in group B. Male/female ratios, average ages, site of insertion, clinical area of insertion and grade of practitioner were similar in both cohorts. There was a significantly higher number of 5-lumen catheters inserted in group B compared with group nB (81% compared with 44%, P < 0.05), reflecting change in hospital practice. More B catheters (51%) were left in situ longer, for 6 to 10 days, compared with nB catheters (31%) (Figure 1). Thirty-one per cent of nB tips grew colonies of at least one pathogen. There was a significant reduction in the number of B tips growing colonies (12% compared with 31%, P < 0.05) (Figure 2). The bundle cohort had no MRSA growth compared with four incidences in the nB group.

Figure 1

Number of days CVCs were left in situ in both cohorts.

Figure 2

Differences in colony growth in the two cohorts. *P < 0.05.

Conclusion

Our results indicate that use of dedicated CVC packs was associated with a significant reduction in the colonisation rate of CVCs, despite lines being left in situ for longer periods and the more frequent use of quinlumen catheters in the intervention group. There was also a trend toward prevention of MRSA colonisation.