A new commercial hub device designed to minimise catheter-related infections was evaluated in a prospective, randomised trial in the intensive care and surgical units of the Hospital de Tortosa Verge de la Cinta in patients in whom the central venous catheters were expected to remain indwelling for at least 7 days. The assessments conducted at catheter withdrawal included cultures of the skin at the catheter site and cultures of the catheter tip and the catheter hubs; moreover, in cases of suspected catheter-related sepsis, samples of peripheral blood and infusion solutions were also cultured. Of the 130 catheters evaluated, 26 (20%) were withdrawn because of suspected catheter-related sepsis; 10 (15%) were in the control group and 16 (24%) in the new product group. Catheter-related sepsis was diagnosed in nine patients, six of whom were in the new product group and three in the control group; all infections in the former group and only one in the latter group were caused by the catheter connection. The rates of catheter hub colonisation (10 cfu) and catheter colonisation (15 cfu in semiquantitative culture and/or >1,000 cfu in quantitative culture) of hub origin were not significantly different between the groups (15 cases in the control group vs. 20 cases in the new product group, and 5 cases in the control group vs. 11 cases in the new product group, respectively). The data indicate that the use of the new catheter hub device is no more effective in preventing catheter-related infection than standard good clinical procedures.