A comparison of tunnelled and nontunnelled subclavian vein catheters: a prospective study of complications during parenteral feeding

Forty-four silastic catheters in 38 surgical patients, nursed in general surgical wards, were inserted under aseptic conditions by the infraclavicular subclavian route. The catheters were randomly allocated to non-tunnelled (NT) (n = 24) or tunnelled (T) (n = 20) groups. Catheters were removed on completion of intravenous feeding or clinical suspicion of catheter infection, and the catheter tip and blood samples taken through the catheter and from the peripheral vein were cultured. There was no significant difference between the two groups in terms of bacteriological infection (defined as two or more cultures of the same organism), clinical infection (defined by elevated temperature returning to normal after catheter removal) and combined infection (when both bacteriological and clinical infection co-existed). When the number of infected catheters was related to the duration of catheter insertion, the incidence of combined catheter related sepsis was reduced with tunnelling (NT: one infected catheter per 35 catheter days, T: one infected catheter per 89 catheter days). This study highlights the risks of subclavian vein catheterisation and emphasises the difficulties in defining catheter sepsis but suggests that its incidence may be reduced if skin tunnelling is employed.