A Controlled Trial of Scheduled Replacement of Central Venous and Pulmonary-Artery Catheters

The incidence of infection increases with the prolonged use of central vascular catheters, but it is unclear whether changing catheters every three days, as some recommend, will reduce the rate of infection. It is also unclear whether it is safer to change a catheter over a guide wire or insert it at a new site. We conducted a controlled trial in adult patients in intensive care units who required central venous or pulmonary-artery catheters for more than three days. Patients were assigned randomly to undergo one of four methods of catheter exchange: replacement every three days either by insertion at a new site (group 1) or by exchange over a guide wire (group 2), or replacement when clinically indicated either by insertion at a new site (group 3) or by exchange over a guide wire (group 4). Of the 160 patients, 5 percent had catheter-related bloodstream infections, 16 percent had catheters that became colonized, and 9 percent had major mechanical complications. The incidence rates (per 1000 days of catheter use) of bloodstream infection were 3 in group 1, 6 in group 2, 2 in group 3, and 3 in group 4; the incidence rates of mechanical complications were 14, 4, 8, and 3, respectively. Patients randomly assigned to guide-wire—assisted exchange were more likely to have bloodstream infection after the first three days of catheterization (6 percent vs. 0, P = 0.06). Insertions at new sites were associated with more mechanical complications (5 percent vs. 1 percent, P = 0.005). Routine replacement of central vascular catheters every three days does not prevent infection. Exchanging catheters with the use of a guide wire increases the risk of bloodstream infection, but replacement involving insertion of catheters at new sites increases the risk of mechanical complications. (N Engl J Med 1992;327: 1062–8.), CENTRAL vascular catheters are widely used in critically ill patients. They permit hemodynamic monitoring and allow access for the administration of fluids, blood products, medications, and total parenteral nutrition, but they also pose a hazard of serious infectious and mechanical complications. Bloodstream infection is a serious catheter-related complication with an estimated case fatality rate of 10 to 20 percent.¹ Single-lumen catheters maintained for parenteral nutrition by strict aseptic techniques have a low cumulative incidence of bacteremia, averaging 6 percent, and are not regarded as requiring routine replacement.^{2 3 4 5 6 7 8} Multilumen and pulmonary-artery catheters used for hemodynamic monitoring in critically ill patients may…