The role of geometric and biomechanical factors in abdominal aortic aneurysm rupture risk assessment

Abstract

The current clinical management of abdominal aortic aneurysm (AAA) disease is based to a great extent on measuring the aneurysm maximum diameter to decide when timely intervention is required. Decades of clinical evidence show that aneurysm diameter is positively associated with the risk of rupture, but other parameters may also play a role in causing or predisposing the AAA to rupture. Geometric factors such as vessel tortuosity, intraluminal thrombus volume, and wall surface area are implicated in the differentiation of ruptured and unruptured AAAs. Biomechanical factors identified by means of computational modeling techniques, such as peak wall stress, have been positively correlated with rupture risk with a higher accuracy and sensitivity than maximum diameter alone. The objective of this review is to examine these factors, which are found to influence AAA disease progression, clinical management and rupture potential, as well as to highlight on-going research by our group in aneurysm modeling and rupture risk assessment.

Figure 1

Aneurysmatic abdominal aorta (left frame is courtesy of University of California at Los Angeles).

Figure 2

Factors affecting evaluation of computational wall mechanics (exclusive of image segmentation errors).

Figure 3

Estimated wall thickness distribution (in mm) as a point cloud resulting from a segmented CT dataset.

Figure 4

1-D size indices computed from segmented CT images: (a) maximum diameter (Dmax), proximal neck diameter (Dneck,p), distal neck diameter (Dneck,d), sac height (Hsac), neck height (Hneck), sac length (Lsac), neck length (Lneck), bulge height (Hb); (b) centroid distance at the maximum diameter (dc).

Figure 5

Model learned by a J48 decision tree, based on highest information gain; L_sac, length of AAA sac; S, surface area; T, tortuosity; , ratio of intraluminal thrombus volume to aneurysm sac volume.

Figure 6

Displacement and stress distribution in three patient-specific geometries for both the CT image based analysis and the zero pressure configuration. Analysis based on the zero pressure configuration yielded a larger peak wall stress (reproduced from unpublished data).

Figure 7

Maximum and average principal stress and strain waveforms for a patient-specific AAA obtained using direct FSI, uncoupled FSI, and transient FEA. The stress and strain follow the inlet velocity waveform rather than the pressure waveform boundary condition (reproduced from unpublished data).

Figure 8

Rationale behind high stresses in saddle shaped surface region; (a) Typical stress distribution obtained by FE analysis under a uniform wall thickness assumption a cut section showing high stresses in the saddle shaped region; (b) Schematic with force directions explaining rationale behind saddle shape and high stress correlation.