Initial clinical experience with an endoluminal spiral prosthesis for treating complicated venous thrombosis and preventing pulmonary embolism.

Fourteen patients with complicated venous thrombosis or recurrent pulmonary embolism were treated by implantation of an endoluminal spiral prosthesis subsequent to balloon angioplasty, surgical thrombectomy or embolectomy, a combination of these, or, in 2 cases, no other treatment. The patients were divided into 2 groups, based on their primary diagnosis and the purpose of the prosthesis. Group I included 8 patients with extensive iliofemoral or caval thrombosis, caused by congenital caval stenosis (1 case) or extravascular compression or retraction (7 cases); 7 of these patients had had previous operations, and the remaining patient had undergone thrombolysis, which failed. The current treatment consisted of balloon angioplasty and surgical thrombectomy or embolectomy, and implantation of an endoluminal spiral stent to prevent elastic recoil of the vessel. In 4 cases, an arteriovenous fistula was constructed and was taken down 3 months later; in 1 additional patient, a bilateral arteriovenous fistula was created. Group II comprised 6 patients with recurrent pulmonary embolism (4 cases), massive pulmonary embolism (1 case), or paradoxical bilateral carotid artery embolism (1 case). Four of these patients underwent surgical thrombectomy or embolectomy, while 2 had no treatment other than filter implantation. All 6 underwent transluminal implantation of a helix caval filter (a modification of the endoluminal spiral stent). All but 1 implantation was accomplished by means of either a transfemoral or a transjugular cutdown; the remaining implantation was performed transatrially after a pulmonary embolectomy. The only device-related complication was a retroperitoneal hematoma in Group I, resulting from perforation of the inferior vena cava by the guidewire during device implantation. This complication necessitated an emergency laparotomy and takedown of the arteriovenous fistula, which resulted in rethrombosis of the left iliofemoral vein. The other 7 stented veins were patent at early phlebographic follow-up, as were all 5 of those studied later. One Group-I patient died 4 months after surgery, due to tumor progression and without signs of caval restenosis. Twelve months postoperatively, 1 Group-II patient died of urosepsis without a recurrence of pulmonary embolism. Four of the 6 Group-II patients were studied late postoperatively, and all of their stented vessels were patent. There was no operative mortality or postoperative embolism. On the basis of these results, we conclude that endoluminal stenting with an expandable spiral prosthesis is a promising method for remote venous reconstruction. Moreover, it appears that the modified stent, or helix caval filter, compares favorably with commercially available filters.