[Percutaneous implant of Denver peritoneovenous shunt: a new opportunity for the interventional radiologist]

The study is aimed at presenting our experience in the implant of Denver peritoneovenous shunts. Medical treatment-resistant ascites, either neoplastic or related to hepatic failure, is highly symptomatic and its treatment is indicated in order to improve patients' quality of life. One of the most efficient methods of treatment consists in implanting a peritoneovenous shunt. The availability of this device and its percutaneous implantation provide Interventional Radiologists with the possibility of expanding their repertoire.

Thirteen shunts were implanted in 12 patients, 10 with neoplastic ascites and 2 with hepatic failure-related ascites. In 1 patient a second device had to be implanted. All the procedures were performed in the Interventional Radiology Department, under local anesthesia and mild sedation. The central venous access was by the subclavian vein in 7 cases and the internal jugular vein in 6 cases. The puncture kit is consists of 2 needles, 1 for venous puncture and 1 for peritoneal puncture, 2 angiographic J-guide wires, 2 peel-away introducers, and a chamber containing the double valve-pump connected with both the venous and the peritoneal catheters. The whole device is placed subcutaneously thus allowing fluids to flow from the peritoneum to the vein either spontaneously, if intra-abdominal pressure exceeds 3cm of water, or by manual compression exerted on the pump itself.

All implants were successfully performed. One transient complication occurred consisting in a mild inflammatory reaction along the subcutaneous catheter route, which promptly solved by antibiotic therapy. So far a total of 1773 catheter/days have been accumulated. 7/10 of the neoplastic patients died from progressive disease after 915 catheter/days (median 120, range 30-180). In a cirrhotic patient the first shunt occluded after 430 days due to hemoperitoneum caused by hepatic biopsy: it was removed and a new one implanted. Five shunts are in now use, with a follow-up of 30, 48, 70, 120 and 160 days each.

The implanting technique was well tolerated by all patients and it could be performed under local anesthesia. The central vein puncture was easy for both accesses but the introducer diameter (12F) and the possibility of clavicle pinch-off induced us to use the internal jugular approach in the last six cases, which provided a reduced risk of pneumothorax and a better catheter track. In the patients with neoplastic ascites we observed neither disease dissemination nor changes in the patients' changed related to the shunt. Our results show that the implant of Denver venous-peritoneal shunts is a relatively easy procedure, which can be performed by Interventional Radiologists on a regular basis.