Catheter-associated bloodstream infection in patients with cancer: comparison of left- and right-sided insertions

Abstract

There is limited research on the relationship between side of insertion of central venous catheter (CVAD) and bloodstream infection risk in patients with cancer. To conduct an exploratory analysis of data from a randomized control trial (RCT) and data from a prospective cohort study to compare infection rates for right- and left-sided insertions. The study populations were patients aged >14 years with cancer from two tertiary hospitals in Brisbane, Australia. The primary endpoint was catheter-associated bloodstream infection (CABSI) adjudicated by blinded assessors. For the RCT, randomized intention-to-treat comparisons were conducted between left- and right-side allocated insertion for early (≤14 days) and late (>14 days) infection using Cox proportional hazards regression. The RCT data were also combined with cohort study data collected from one of the hospitals prior to the RCT and non-randomized comparisons conducted between left- and right-sided insertions. In 634 randomly allocated CVADs there were 141 CABSIs. Analysis showed strong evidence of right-side allocated insertions having an increased risk of early infection by 2.5 times (95% confidence interval (CI): 1.3-4.7); however, there was no evidence of increased risk for late infection (hazard ratio: 1.06; 95% CI: 0.71-1.59). Results from analysis of the RCT and cohort study data combined (2786 CVADs and 385 CABSIs) were similar. There appears to be an increased risk of CABSI in patients with cancer for CVAD inserted into the right-side for around two weeks after line insertion. The mechanism underpinning the increased risk is unknown.