Thrombus Prevention in Tunneled Central Venous Catheters

The purpose of this study is to evaluate the effect of scheduled urokinase cleansing on the incidence of withdrawal occlusion (able to flush catheter, but unable to withdraw blood) and thrombus formation in tunneled central venous catheters (CVCs). These problems affect the utilization of these catheters and may even necessitate their removal. Thrombus formation in and around the tunneled CVC is felt to occur as a result of blood vessel wall injury caused by the CVC (1). Obstruction of the catheter tip by thrombus is felt to be the most frequent cause of withdrawal occlusion and occurs in 30% of tunneled CVCs (2,3,4). To treat thrombus buildup, investigators have attempted varying schedules of urokinase to "declot" or lyse the fibrin clot. One of the earliest studies published reported a 98.6% success rate in clearing occluded catheters (5). This study helped define the role of urokinase in catheter declotting. Today, urokinase declotting is a common and, unfortunately, frequent necessity for maintaining the patency of tunneled CVCs. Most CVCs require declotting with urokinase as a response to catheter and/or withdrawal occlusion. In a small study at MD Anderson Hospital, 10 catheters with withdrawal occlusion were treated with urokinase to "cleanse" them of thrombus formation with good results. These catheters remained cleared with prophylactic urokinase (6). These results led the investigator to recommend studying the use of a prophylactic schedule of urokinase to decrease withdrawal occlusion in CVCs. Based on these study results and our knowledge of thrombus formation around CVCs, this study plans to use urokinase on a schedule designed to prevent or treat early thrombus formation around the CVC. The study design is a prospective randomized clinical trial utilizing a convenience sample of oncology patients requiring tunneled CVCs. A total of 130 consecutive patients (2 groups of 65) will be enrolled over a 2 year period. The control group will receive no scheduled urokinase; the experimental group will receive scheduled urokinase cleansing. Data analysis will include the use of chi-square methods to detect a reduction in overall withdrawal occlusion and analyses of variance to compare CVC tip clot sizes obtained via catheter dye studies. Kruskal-Wallis non-parametric statistics will be used to detect other differences between the two groups.