P V Petrik1, W S Moore. 1. Division of Vascular Surgery, UCLA Center for the Health Sciences, Los Angeles, CA, USA.
Abstract
OBJECTIVE: The failure to maintain a secure exclusion of aortic aneurysms with intraluminally placed grafts has been termed endoleak. We performed a retrospective review of our first 100 transluminally repaired abdominal aortic aneurysms (AAAs) in an effort to identify preoperative factors that could predict which patients would have endoleaks. METHODS: Between February 1993 and September 1998, 100 infrarenal aneurysms were treated with tube (39), bifurcated (45), and aortoiliac grafts (16). Endoleaks (early and late) developed in 34 patients. Preoperative computed tomography scans and angiograms for all patients were individually inspected by a single reviewer. Aortic characteristics analyzed included number of patent lumbar arteries, presence of a patent inferior mesenteric artery (IMA), calcification and thrombus at proximal and distal attachment sites, proximal aortic angulation, and graft-vessel size discrepancy at proximal and distal attachment sites. The prevalence of the preoperative factors was compared among patients with and without endoleaks. RESULTS: Endoleaks developed in 44% of tube, 33% of bifurcated, and 47% of aortoiliac grafts (P =.51). Correlation between total number of patent lumbar arteries, presence of a patent IMA, and endoleaks was not significant (P =.44,.95). Calcification at either proximal or distal attachment site did not increase the risk of endoleaks (P =.50,.62). The presence of thrombus at the attachment site (proximally or distally) also failed to increase endoleak rates (P =.12,.78). Degree of proximal aortic angulation did not differ between groups (P =.39). Size discrepancies between graft and aorta or iliac vessels at proximal or distal sites did not significantly differ (P >.54, >.13). Subgroup analysis of endoleaks with different tube types also failed to demonstrate significant differences among the three graft types (P >.05). CONCLUSION: Endoleaks develop in a significant number of endovascularly repaired AAAs. We were unable to demonstrate a statistically significant association with anatomic characteristics thought to predispose to the development of endoleaks. We find no predictive value associated with these anatomic factors.
OBJECTIVE: The failure to maintain a secure exclusion of aortic aneurysms with intraluminally placed grafts has been termed endoleak. We performed a retrospective review of our first 100 transluminally repaired abdominal aortic aneurysms (AAAs) in an effort to identify preoperative factors that could predict which patients would have endoleaks. METHODS: Between February 1993 and September 1998, 100 infrarenal aneurysms were treated with tube (39), bifurcated (45), and aortoiliac grafts (16). Endoleaks (early and late) developed in 34 patients. Preoperative computed tomography scans and angiograms for all patients were individually inspected by a single reviewer. Aortic characteristics analyzed included number of patent lumbar arteries, presence of a patent inferior mesenteric artery (IMA), calcification and thrombus at proximal and distal attachment sites, proximal aortic angulation, and graft-vessel size discrepancy at proximal and distal attachment sites. The prevalence of the preoperative factors was compared among patients with and without endoleaks. RESULTS: Endoleaks developed in 44% of tube, 33% of bifurcated, and 47% of aortoiliac grafts (P =.51). Correlation between total number of patent lumbar arteries, presence of a patent IMA, and endoleaks was not significant (P =.44,.95). Calcification at either proximal or distal attachment site did not increase the risk of endoleaks (P =.50,.62). The presence of thrombus at the attachment site (proximally or distally) also failed to increase endoleak rates (P =.12,.78). Degree of proximal aortic angulation did not differ between groups (P =.39). Size discrepancies between graft and aorta or iliac vessels at proximal or distal sites did not significantly differ (P >.54, >.13). Subgroup analysis of endoleaks with different tube types also failed to demonstrate significant differences among the three graft types (P >.05). CONCLUSION: Endoleaks develop in a significant number of endovascularly repaired AAAs. We were unable to demonstrate a statistically significant association with anatomic characteristics thought to predispose to the development of endoleaks. We find no predictive value associated with these anatomic factors.