Biomechanical rupture risk assessment of abdominal aortic aneurysms using clinical data: A patient-specific, probabilistic framework and comparative case-control study

Abstract

We present a data-informed, highly personalized, probabilistic approach for the quantification of abdominal aortic aneurysm (AAA) rupture risk. Our novel framework builds upon a comprehensive database of tensile test results that were carried out on 305 AAA tissue samples from 139 patients, as well as corresponding non-invasively and clinically accessible patient-specific data. Based on this, a multivariate regression model is created to obtain a probabilistic description of personalized vessel wall properties associated with a prospective AAA patient. We formulate a probabilistic rupture risk index that consistently incorporates the available statistical information and generalizes existing approaches. For the efficient evaluation of this index, a flexible Kriging-based surrogate model with an active training process is proposed. In a case-control study, the methodology is applied on a total of 36 retrospective, diameter matched asymptomatic (group 1, n = 18) and known symptomatic/ruptured (group 2, n = 18) cohort of AAA patients. Finally, we show its efficacy to discriminate between the two groups and demonstrate competitive performance in comparison to existing deterministic and probabilistic biomechanical indices.

On stress in abdominal aortic aneurysm: Linear versus non‐linear analysis and aneurysm rupture risk

“…This may have important implications in the models that calculate rupture risk locally. 7,8,10 The observed differences between the results obtained using BioPARR and Non-LISA are consistent with the results by Man et al 24 In their study, Non-LISA nonlinear iterative approach was used to compare the stress fields in AAA predicted using non-linear material with very low initial stiffness and linear material.…”

“…To address shortcomings of the maximum diameter-based assessment of AAA rupture risk, biomechanical analysis has been proposed. [7][8][9][10][11][12] In the biomechanical analysis, an artery rupture is treated as a mechanical event that occurs when the local stress in artery wall exceeds the local strength of the artery wall. It has been suggested in several studies that patients with AAA symptoms will have high stress in the arterial wall of AAA and, consequently, be at higher risk of AAA rupture.…”

Patient‐specific computational modeling of endovascular aneurysm repair: State of the art and future directions

“…Aneurysms and dissections need to be repaired to avoid aortic rupture, an almost always fatal event. Aortic rupture occurs when aortic wall stress exceeds aortic wall strength, 61–65,68,71,72,98 which is the stress at which the wall cannot longer withstand the forces applied to it, for instance forces from blood pressure acting on the luminal surface of the aortic wall. Rupture represents the main concern associated with TAA and AAA, as well as AD because rupture carries high rates of mortality (80%–90%), considering that most patients do not even make it to an aortic center on time 99–104 .…”

“…An aneurysm, however, results from aortic wall tissue degradation and inflammation, that alter the mechanical properties of the aortic wall in ways that vary locally in the same patient, and are different from patient to patient. These mechanical properties depend on the degree of tissue degradation, which is unknown in an individual patient 61,71,73,108,125 . Moreover, the aortic wall may have calcifications that further alter its mechanical properties locally and globally.…”

“…As mentioned before, patient‐specific aortic properties, such as the aortic wall thickness and the aortic wall stiffness and collagen dispersion, are usually difficult to estimate and therefore mostly treated in terms of population‐averaged quantities instead of patient‐specific ones. Some studies address these aortic uncertainties with probabilistic approaches 49,61,71 in the context of patient‐specific AAA rupture prediction. Incorporating these probabilistic approaches into computational EVAR, however, would drastically increase the computational effort.…”