Catheter-Related Bloodstream Infection in Jugular Versus Subclavian Central Catheterization
I replicated their analysis (RevMan version 5.3; the Nordic Cochrane Centre, The Cochrane Collaboration, Copenhagen, Denmark, 2014) with the following potential improvements: 1) The generic inverse variance was used instead of the dichotomous data type, to incorporate catheter-days and compute the number of CRBSI/catheter-days in the jugular group divided by the same ratio in the subclavian group; 2) Two large (2, 3), high quality, randomized controlled trials for other outcomes reporting CRBSI by CVC site were added; 3) The number of CRBSI in the jugular group in the study by Arvaniti et al (4) was corrected from 0 to 1 (consistent with the jugular vs femoral comparison), since seven CRBSIs occurred in their study (4), five in subclavian, and one in femoral (1). The results are shown in Figure 1.
The occurrence of CRBSI was higher for jugular when compared with subclavian (risk ratio, 1.80 [95% CI, 1.01–3.20]; I2 = 50%), but the results were heterogeneous and driven for 86.4% by observational studies. Nevertheless, this analysis suggests that the subclavian site may be associated with a lower risk of CRBSI.
More importantly, the 3SITES trial (5) demonstrated that the risk of severe intravascular complications such as CRBSI or symptomatic deep-vein thrombosis was significantly higher in the jugular (hazard ratio, 2.1 [95% CI, 1.0–4.3]; p = 0.04) when compared with the subclavian site, in the intent to treat dataset. This evidence should be kept in mind when selecting the “convenient” jugular route, among patients without contraindication to subclavian catheterization at risk of intravascular complications.