Objective: Late complications and graft failures have recently cast serious doubts on the durability of endovascular repair of abdominal aortic aneurysms (AAA). The results of a multicenter trial comparing a bifurcated endograft (AB) with standard open repair (OR) were reviewed to assess the late findings of both methods of AAA treatment.
Patients and methods: In a multicenter study of AB versus OR conducted from December 1995 to February 1998, 242 patients with AAA successfully treated with an AB and 111 control patients treated concurrently with OR were followed up at least yearly. Twenty-five immediate conversions were excluded from late follow-up. All imaging modalities obtained during follow-up were reviewed by a core laboratory for AAA size, endoleaks, migration, and device integrity. Clinical outcomes at the yearly visits were compared. All death reports were reviewed to classify the cause of death.
Results: Average follow-up for the AB group was 36 months, with 194 patients at 3 years and 55 patients at 4 years. The cumulative mortality rate was similar between the AB (15.7%) and OR groups (12.6%; P =.59). The significant early benefit to the AB group in cardiopulmonary complications was no longer evident by 3 years. However, the AB advantage in total and bowel complications, as well as the higher renal complication rates, persisted. At 3 years, 73.7% of patients showed a significant reduction of their AAA size, whereas 25.7% still had an endoleak. One migration and two single hook fractures were noted. Graftrelated reinterventions were performed in 50 patients (20%) without any deaths. Twenty-eight patients (11.6%) underwent interventions for limb flow compromise, whereas 25 were treated for endoleak. Late conversion to OR was required in five patients (2%). No AAA ruptures were encountered in either group.
Conclusions: Rupture-free survival rates after treatment of AAA with the bifurcated AB are similar to those of the OR group. Notably the proximal attachment system is relatively stable and the AAA shrinks in three of four patients treated. Reinterventions are nonetheless required in nearly one of five patients. Although most late procedures are percutaneous, counseling regarding possible future interventions is necessary.