Endoleaks after endovascular graft treatment of aortic aneurysms: Classification, risk factors, and outcome

Purpose: Incomplete endovascular graft exclusion of an abdominal aortic aneurysm results in an endoleak. To better understand the pathogenesis, significance, and fate of endoleaks, we analyzed our experience with endovascular aneurysm repair. Methods: Between November 1992 and May 1997, 47 aneurysms were treated. In a phase I study, patients received either an endovascular aortoaortic graft (11) or an aortoiliac, femorofemoral graft (8). In phase II, procedures and grafts were modified to include aortofemoral, femorofemoral grafts (28) that were inserted with juxtarenal proximal stents, sutured endovascular distal anastomoses within the femoral artery, and hypogastric artery coil embolization. Endoleaks were detected by arteriogram, computed tomographic scan, or duplex ultrasound. Classification systems to describe anatomic, chronologic, and physiologic endoleak features were developed, and aortic characteristics mere correlated with endoleak incidence. Results: Endoleaks were discovered in 11 phase I patients (58%) and only six phase II patients (21%; p < 0.05). Aneurysm neck lengths 2 cm or less increased the incidence of endoleaks (p < 0.05). Although not significant, aneurysms with patent side branches or severe neck calcification had a higher rate of endoleaks than those without these features (47% vs 29% and 57% vs 33%, respectively), and patients with iliac artery occlusive disease had a lower rate of endoleaks than those without occlusive disease (18% vs 42%). Endoleak classifications revealed that most endoleaks were immediate, without outflow, and persistent (71% each), proximal (59%), and had aortic inflow (88%). One patient with a persistent endoleak had aneurysm rupture and died. Conclusions: Endoleaks complicate a significant number of endovascular abdominal aortic aneurysm repairs and may permit aneurysm growth and rupture. The type of graft used, the technique of graft insertion, and aortic anatomic features all affect the rate of endoleaks. Anatomic, chronologic, and physiologic classifications can facilitate endoleak reporting and improve understanding of their pathogenesis, significance, and fate. (J Vasc Surg 1998;27:69-80.).