Central venous catheter tip position and malfunction in a paediatric oncology unit.

An audit of 151 central venous catheters (CVCs) in 118 children with malignant disease was carried out over 20 months. The types included 31 valved silastic (Groshong), 58 non-valved silastic (Hickman), and 62 non-valved polyurethane (Cuff Cath) CVCs. There was no difference between the three groups with regard to the clinical diagnosis. The mean patient age at catheter insertion was 5.5 years and the mean weight 21.6 kg. None of the catheter types were associated with an increased risk of problems at insertion, migration, mechanical damage, blockage, sampling, or catheter infection. The incidence of catheter infection was 1.4/1,000 catheter days. Exit-site infection was less frequent with Groshong CVCs (P <0.05), which were in situ for the shortest period. The risk of problems with blood sampling was significantly increased in those catheters whose tip was sited outside the right atrium (P <0.005). For the 60 CVCs removed electively, the mean duration in situ was similar for all catheter types; 43 were removed following a problem. Of these, Groshong catheters were in situ for the shortest period (P = 0.05), probably as a result of delayed anchoring of the cuff. The tip position was the single most important determinant in the correct functioning of CVCs, irrespective of the type of catheter. Intraoperative screening of the tip position at catheter insertion is therefore mandatory for optimal catheter functioning.