Central venous access sites for the prevention of venous thrombosis, stenosis and infection in patients requiring long-term intravenous therapy

Central venous access (CVA), in which a large bore catheter is routed through a vein in the neck, upper chest or femoral area, is needed to give drugs that cannot be given by mouth or via a conventional cannula in the arm. To establish whether either the jugular, subclavian or femoral CVA routes result in a lower incidence of venous thrombosis, venous stenosis or infection related to CVA devices.To determine whether the circumference of a long-term central venous access device influences the incidence of venous thrombosis, venous stenosis or infection related to CVA devices. We searched the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library 2006, Issue 4), MEDLINE, CINAHL, EMBASE (from inception to December 2006), reference lists of identified trials, and bibliographies of published reviews. We also contacted researchers in the field. There were no language restrictions. We included randomized controlled trials comparing central venous catheter insertion routes. Two authors assessed potentially relevant studies. We resolved disagreements by discussion. Relevant outcomes were: venous thrombosis, venous stenosis, infection related to CVA devices, mechanical complications (e.g misplaced catheter, minor bleeding, haematoma). We considered 83 studies for inclusion in the review. Six studies appeared eligible but five were subsequently excluded because they did not randomize participants for either site of access or catheter circumference size. One study was a high quality block randomized controlled trial. Allocation concealment was good and randomization was by a central computer. In all, 293 patients were randomized to a femoral or a subclavian CVA group. Results from this one trial were as follows. 1. CATHETER-RELATED INFECTIOUS COMPLICATIONS: Infectious complication (colonization with or without sepsis: the relative risk (RR) was 4.57 (95% confidence interval (CI) 1.95 to 10.71) favouring subclavian over femoral access. Major infectious complications (sepsis with or without bacteremia): the RR was 3.04 (95% CI 0.63 to 14.82) favouring subclavian access. Colonized catheter (greater than 103 colony-forming units/mL of gram positive microorganisms): the RR was 3.65 (95%CI 1.40 to 9.56) favouring subclavian access. Colonized catheter (greater than 103 colony-forming units/mL of gram negative microorganisms): the RR was 5.41 (95% CI 1.61 to 18.15) favouring subclavian access. 2. CATHETER-RELATED MECHANICAL COMPLICATIONS: Overall compl