Central venous device infections are associated with increased physical and psychological morbidity, mortality, length of stay, and costs. The aim of this study was to prove the efficacy of pulsatile flushing to prevent the bacterial colonization of vascular access devices. One hundred and forty four tests using 576 polyurethane short venous access catheters were performed. Four catheters per test were polluted with a fibronectin-serum albumin solution. Three were filled with a Staphylococcus aureus broth; one served as negative control. One contaminated catheter was not flushed (positive control), and two were flushed (10 mL.sec−1) with normal saline solution, either by ten successive boluses of 1 mL each or by one bolus of 10 mL. Each catheter was cultivated. The S. aureus quantity observed after continuous flushing was significantly higher than that observed after pulsative flushing (P<0.001). Unflushed catheters were 20.71 and 6.42 times more polluted than catheters flushed with the pulsative method or the continuous method, respectively. Pulsative flushing was at least twice as effective as continuous flushing in reducing the S. aureus count. Pulsative flushing is more effective than continuous flushing in reducing the endoluminal contamination. Pulsative flushing is a simple, effective, and inexpensive technique to reduce catheter bacterial colonization.