Preliminary 3D computational analysis of the relationship between aortic displacement force and direction of endograft movement

Objective:Endograft migration is usually described as a downward displacement of the endograft with respect to the renal arteries. However, change in endograft position is actually a complex process in three-dimensional (3D) space. Currently, there are no established techniques to define such positional changes over time. The purpose of this study is to determine whether the direction of aortic endograft movement as observed in follow-up computed tomography (CT) scans is related to the directional displacement force acting on the endograft. Methods: We quantitated the 3D positional change over time of five abdominal endografts by determining the endograft centroid at baseline (postoperative scan) and on follow-up CT scans. The time interval between CT scans for the 5 patients ranged from 8 months to 8 years. We then used 3D image segmentation and computational fluid dynamics (CFD) techniques to quantitate the pulsatile displacement force (in Newtons [NI]) acting on the endografts in the postoperative configurations. Finally, we calculated a correlation metric between the direction of the displacement force vector and the endograft movement by computing the cosine of the angle of these two vectors. Results: The average 3D movement of the endograft centroid was 18 mm (range, 9-29 mm) with greater movement in patients with longer follow-up times. In all cases, the movement of the endograft had significant components in all three spatial directions: Two of the endografts had the largest component of movement in the transverse direction, whereas three endografts had the largest component of movement in the axial direction. The magnitude and orientation of the endograft displacement force varied depending on aortic angulation and hemodynamic conditions. The average magnitude of displacement force for all endografts was 5.8 N (range, 3.7-9.5 N). The orientation of displacement force was in general perpendicular to the greatest curvature of the endograft. The average correlation metric, defined as the cosine of the angle between the displacement force and the endograft centroid movement, was 0.38 (range, 0.08-0.66). Conclusions: Computational methods applied to patient-specific postoperative image data can be used to quantitate 3D displacement force and movement of endografts over time. It appears that endograft movement is related to the magnitude and direction of the displacement force acting on aortic endografts. These methods can be used to increase our understanding of clinical endograft migration. (J Vase Surg 2010;51:1488-97.)