Prevention of vascular catheter-related infection remains an important priority. This review focuses on salient controversies regarding optimal preventive methods. Intensity of surveillance for nosocomial infections was the single most important predictor of prevention in the Study of the Efficacy of Nosocomial Infection Control (SENIC). Used suboptimally by most hospitals in the SENIC study, surveillance is probably conducted even less today. There has been one randomized trial of the optimal method of aseptic insertion for central venous catheters and none comparing the 2 most frequently used sites. Scheduled replacement did not prevent infection in multiple randomized trials but, according to a recent survey, was still being used frequently. Chlorhexidine preparation of skin before and during catheterization has significantly reduced colonization of catheters in multiple randomized trials and should be used. Impregnation of catheter and/or hub surfaces with antiseptics raises less concern about fostering the development of antibiotic resistance than does the use of antibiotics for this purpose.