Polar orientation of renal grafts within the proximal seal zone affects risk of early type IA endoleaks after chimney endovascular aneurysm repair.

Document Type


Publication Date



Aged; Aged, 80 and over; Aortic Aneurysm, Abdominal; Aortography; Blood Vessel Prosthesis; Blood Vessel Prosthesis Implantation; Chi-Square Distribution; Computed Tomography Angiography; Endoleak; Endovascular Procedures; Female; Humans; Kaplan-Meier Estimate; Logistic Models; Male; Middle Aged; Multivariate Analysis; Odds Ratio; Prosthesis Design; Renal Artery; Retrospective Studies; Risk Factors; Stents; Time Factors; Treatment Outcome; cards


OBJECTIVE: The objective of this study was to describe the polar orientation of renal chimney grafts within the proximal seal zone and to determine whether graft orientation is associated with early type IA endoleak or renal graft compression after chimney endovascular aneurysm repair (ch-EVAR).

METHODS: Patients who underwent ch-EVAR with at least one renal chimney graft from 2009 to 2015 were included in this analysis. Centerline three-dimensional reconstructions were used to analyze postoperative computed tomography scans. The 12-o'clock polar position was set at the takeoff of the superior mesenteric artery. Relative polar positions of chimney grafts were recorded at the level of the renal artery ostium, at the mid-seal zone, and at the proximal edge of the graft fabric. Early type IA endoleaks were defined as evidence of a perigraft flow channel within the proximal seal zone.

RESULTS: There were 62 consecutive patients who underwent ch-EVAR (35 double renal, 27 single renal) for juxtarenal abdominal aortic aneurysms with a mean follow-up of 31.2 months; 18 (29%) early type IA "gutter" endoleaks were identified. During follow-up, the majority of these (n = 13; 72%) resolved without intervention, whereas two patients required reintervention (3.3%). Estimated renal graft patency was 88.9% at 60 months. Left renal chimney grafts were most commonly at the 3-o'clock position (51.1%) at the ostium, traversing posteriorly to the 5- to 7-o'clock positions (55.5%) at the fabric edge. Right renal chimney grafts started most commonly at the 9-o'clock position (n = 17; 33.3%) and tended to traverse both anteriorly (11 to 1 o'clock; 39.2%) and posteriorly (5 to 7 o'clock; 29.4%) at the fabric edge. In the polar plane, the majority of renal chimney grafts (n = 83; 85.6%) traversed90 degrees were independently associated with early type IA endoleaks (odds ratio, 11.5; 95% confidence interval, 2.1-64.8) even after controlling for other device and anatomic variables. Polar orientation of the chimney grafts was not associated with graft kinking or compression (P = .38) or occlusion (P = .10). Takeoff angle of the renal arteries was the most significant predictor of chimney graft orientation. Caudally directed arteries (takeoff angle >30 degrees) were less likely to have implanted chimney grafts that traversed >90 degrees in polar angle (odds ratio, 0.09; 95% confidence interval, 0.01-0.55).

CONCLUSIONS: Renal chimney grafts vary considerably in both starting position and their polar trajectory within the proximal seal zone. Grafts that traverse >90 degrees in polar angle within the seal zone may be at increased risk of early type IA endoleaks and require more frequent imaging surveillance. Caudally directed renal arteries result in a more favorable polar geometry (eg, cranial-caudal orientation) with respect to endoleak risk and thus are more ideal candidates for parallel graft strategies.

Clinical Institute

Cardiovascular (Heart)




Center for Cardiovascular Analytics, Research + Data Science (CARDS)


Brant Ullery is affiliated with Providence St. Joseph Health.