Optimal Technique for Imaging Iliac Segments During Endovascular Repair of Abdominal Aortic Aneurysms

Abstract

To determine if oblique angulation of the image intensifier is adequate to image the entire length of the common iliac artery during endovascular aneurysm repair or if additional caudal tilt is necessary. Using a 3D workstation, the apparent level of the iliac bifurcation (distal limit of the stent-graft) was determined on computed tomographic angiography by profiling the common iliac segment in oblique angulation only and repeated with a combination of oblique angulation and caudal tilt. Two independent observers measured twice the apparent length of the iliac segment in both profiles for 50 patients according to a set protocol. Intra- and interobserver variability was calculated using the Bland and Altman method; the differences between the two different profiles were tested using paired t tests. Of the 50 CTA datasets reviewed, 2 datasets were excluded owing to extensive calcification of the iliac system that prevented accurate interpretation of the image. Of the 96 segments studied, the iliac segments appeared longer (better profiled) with a combination of caudal tilt and oblique angulation in 80%, with an average discrepancy of 9 mm for observer 1 (range -1 to +28) and 7 mm for observer 2 (0 to +26). The effect of caudal tilt was statistically significant for individual observers (p = 0.001 and 0.024, respectively). Forty-six percent of iliac segments measured by observer 1 and 35% by observer 2 showed that the addition of caudal tilt resulted in improved profiling by at least 10 mm. Although inter- and intraobserver variation was significant, the gain in apparent iliac length with the addition of caudal tilt was preserved. When profiled with oblique angulation alone, the location of the iliac bifurcation may appear higher than its true location, resulting in underutilization of the iliac segment by >10 mm in over a third of the patients. The problem is corrected by employing additional caudal tilt.