Background: Thoracic aortic aneurysms (TAAs) and type B aortic dissections (ADs) are relatively frequent, serious conditions that are often managed nonoperatively because of perceived poor outcome of standard surgical reconstruction. Recently developed stent graft techniques represent a more attractive, less invasive option. We sought to determine the technical feasibility and safety of endovascular repair in the thoracic aorta with a retrospective review of our experience with such an approach. Methods. Forty-seven patients received thoracic stent graft implants during the 4-year period ending March 31, 2002. All patients signed an Institutional Review Board-approved informed consent. Thirty-one patients had TAAs, and 16 had ADs. Device design and implant strategy were on the basis of evaluation of morphology with angiography and computed tomographic scan. The procedures were done with fluoroscopic guidance, with local anesthesia in five cases, spinal anesthesia in 19 cases, and general anesthesia in 23 cases. Endovascular access was achieved with femoral cutdown in 41 cases and a temporary iliac conduit in six cases. A Talent patient-specific device, with 4-mm to 6-mm oversize, was used in all. Proximal endograft attachment was in the descending thoracic aorta in 16 cases, parasubclavian in 21 cases, and the suprasubclavian aorta in 10 cases. Eight patients had adjunctive cervical reconstruction to transpose or revascularize the left subclavian or left common carotid arteries, enabling more proximal endograft attachment in the aortic arch. Results. Access failure occurred in one patient (2.1%). One patient (2.1%) died within 30 days of access-related iliac artery rupture. Another death occurred at 60 days from a ruptured thoracoabdominal aneurysm with type I endoleak. No instances of paraplegia, stroke, or surgical conversion were seen. Five patients (TAA) were found to have endoleak on 30-day computed tomographic scan. Repair of type I endoleak was undertaken in three cases at 1, 4, and 6 months. Eight patients (17%) had adverse events within the first 3 0 days. Length of follow-up ranged from 1 to 44 months, with a mean of 18 months. Two patients were lost to follow-up, and one withdrew from the study. Four additional mortalities were observed, none related to the endograft or aortic pathology. Conclusion: Stent graft repair of TAA and AD is feasible and can be achieved with technical success and relatively low rates of perioperative morbidity and mortality. The Talent customized design proved versatile in various morphologies. More information is needed on indications, clinical efficacy, and long-term results.
