Morphology and mechanical properties of abdominal aortic aneurysms

Abstract

Abdominal Aortic Aneurysm (AAA) rupture is a life-threatening event, but can, when diagnosed timely, be avoided by either endovascular or conventional repair. Since these methods are accompanied by a high mortality rate, the decision to operate should only be made when the rupture risk exceeds the risks of repair. To date, the maximum diameter of the aneurysm is used as a measure for rupture risk and hence as a criterion for surgical intervention. It has, however, been shown already that small aneurysms do rupture sometimes, while some large aneurysms have not ruptured yet, and therefore the diameter criterion alone may not be suf??cient. Since rupture of an AAA occurs when locally the wall stress exceeds the strength of the vessel wall, it is generally believed that wall stress distribution could help to better assess AAA rupture. Patient speci??c models for wall stress computations do not only require the geometry of the aneurysm but also the mechanical properties of the aneurym tissue. Studies on the material properties of vessel wall have mainly focused on describing AAA vessel wall and thrombus as a homogeneous material. Local inhomogeneities can have large in??uences on the stress magnitude and distributions. AAA vessel wall may contain inhomogeneities such as calci??cations and atherosclerotic plaques. Furthermore, the thrombus, which is present in the majority of the aneurysms may in??uence the local mechanics considerably. It is a layered ??brin structure, that shows different levels of degeneration. The level of degeneration of the ??brin structure might have in??uence on the mechanical properties of the thrombus. To improve future wall rupture risk prediction based on wall stress, the objective of this study is to obtain the local mechanical properties of both thrombus and the AAA vessel wall. AAA vessel wall and thrombus are obtained from patients treated with conventional surgery. The viscoelastic behaviour of thrombus is determined using plate-plate rheomix etry. To study the changes in mechanical properties throughout the thickness of the thrombus a radially oriented stack of samples was used. In the small strain regime frequency sweep test are peformed and the elastic and viscous moduli are found to be in the range of 1.7 ?? 1.3 kPa and 0.2 ?? 0.1 kPa respectively. Since large deformations occur in thrombus the non-linear properties are determined by stress relaxation experiments. To describe the phenomena observed experimentally, a non-linear multi mode model is used. The parameters for this model are obtained by ??tting this model successfully to the experiments in both the linear and non-linear regime. To determine the morpholgy of the AAA vessel wall, the applicability of multi contrast MRI to discriminate the components of the vessel wall is studied. Multi contrast MR results are compared to the golden standards, histology and ??CT. Components like the media, calci??ed deposits, areas containing cholesterol, thrombus and the adventitia containing fat cells and vasa vasorum can be recognised in the MR images. The results obtained so far are not suitable for an automatic classi??cation by an unsupervised clustering algorithm. To obtain the mechanical properties of the components present in the vessel wall a mixed numerical experimental method is proposed. This method is a robust way to determine the mechanical properties of the individual components without having to isolate them. The method is applied to AAA vessel wall samples. The morphology is obtained with multi contrast MR. The Young's moduli of media, adventitia, calci??cation, an area with cholesterol crystals and thrombus are determined. Although the media is less stiff than the adventitia the moduli are in same the order of magnitude (?? 1.0 MPa). The stiffness of thrombus is found to be an order of magnitude lower (?? 40 kPa). An area with a lot of cholesterol is stiffer than the rest of the wall (?? 7.5 MPa) and a calci??ed area is even stiffer (?? 50 MPa). By implementing the ??ndings for thrombus in a patient speci??c model it is shown that thrombus does change the stress distribution and peak wall stresses, but the effect is much lower than reported in literature by others. The presence of calci??cation has a large effect on the wall stress distribution. It is shown that the wall stress distribution is in??uenced by the accuracy with which the geometry of the calci??cation is decribed. The limited resolution available in the clinical settting results in a non-realistic stress distribution. For correct implementation of inhomogeneities of the vessel wall the spatial resolution of both imaging techniques and computations should increase. The contribution of this work consists not only of the reported morphologies and mechanical properties. The observations reported in this work also suggest that effort should be put into development of clinical methods to include the morphology and mechanical properties of the aneurysm into the rupture risk analysis.