[Insertion and management of long-term central venous devices: role of radiologic imaging techniques].

INTRODUCTION: Anticancer chemotherapy causes irreversible damage to the endothelial wall of small vessels. This is the reason why long-term (more than 3 months) central venous devices are essential to administer chemotherapy drugs to cancer patients and antibiotics for chronic or severe infections and in patients requiring long-term parenteral nutrition. We report our experience with the percutaneous implantation of central venous devices in a radiology department.

MATERIAL AND METHODS: March, 1993, to August, 1997, eighty-seven consecutive patients (26 men and 61 women, mean age: 55 years) were examined. The indications for central venous catheter placement included anticancer chemotherapy in 82 cancer patients, repeated blood transfusion in one patient with bone marrow aplasia and nutritional support in four cancer patients. Eighty-four central venous devices (75 totally subcutaneous systems--Port-a-cath Dome--, and 9 partially tunneled catheters--Groshong) were inserted. The average follow-up was 6.5 months (range: 1-18). All procedures were performed in the radiology department and venous access was achieved with fluoroscopy using the Seldinger technique. Chest radiography with the patient standing was routinely performed after the procedure and repeated the day after to assess the catheter position and the presence of pneumothorax. The venous catheters were placed in the subclavian vein in 68 cases (12 in the right side and 56 in the left side), internal jugular vein in 12 cases (9 in the right side and 3 in the left side) and right femoral vein in 4 cases. We prefer the subclavian vein (80.9%) for better cosmetic results, wider catheter angulation and easier fixation to the deep plane.

RESULTS: The first access failed in 6 cases (6.8%). A pneumothorax occurred in 4 patients (4.7%) and late complications were seen in 15 patients (17.8%) after a mean of 15.7 weeks (range: 2-48). Catheter-related infections developed in 6 patients (7.1%) after a mean of 20 weeks (range: 5-48). The microorganisms cultured from these catheters was the Staphylococcus epidermidis. After two weeks' specific antibiotic therapy, all the devices were removed. Deep venous thrombosis occurred only in one patient after 10 months and was successfully treated with direct thrombolytic infusion. The catheter was displaced in the right atrium in two patients after 11 and 12 weeks, respectively: both catheters were removed by transfemoral catheterization.

CONCLUSIONS: The percutaneous implantation of--long-term central venous devices is a safe and tolerable procedure. In our experience, the radiology-assisted placement of these devices offers many advantages over surgical implantation. In particular, fluoroscopy allows direct visualization of the catheter position while insertion and positioning are essentially "blind" at surgery, which complicates venous access and increases the risk of catheter malpositioning. Radiologic follow-up is also useful to depict and correct complications.