Geometric, biomechanical and molecular analyses of abdominal aortic aneurysm

Abstract

Background: Abdominal aortic aneurysm (AAA) is defined as a dilatation of the abdominal aorta of 30 mm in diameter or more. Main risk factors are smoking, age and male sex. Pathophysiological features include inflammation, smooth muscle cell loss and destruction of the extracellular matrix. The AAA is typically asymptomatic but can expand and eventually rupture, with a mortality of 70-80% as a result. Risk factors for rupture include a large diameter, female sex, active smoking, high blood pressure and low body mass index (BMI). There is no medical treatment to inhibit growth or rupture of AAA. The only measure to prevent rupture in a large AAA is aortic surgery. This intervention carries its own significant risk of morbidity and mortality, necessitating a risk stratification method. The diameter is currently used to decide when to operate on an AAA and it is repeatedly monitored until the threshold for surgery is reached. However, this measurement leaves room for improvement, as the individual aneurysm growth rate is difficult to predict and some large AAAs do not rupture while in other patients, small AAAs rupture during surveillance. Finite element analysis (FEA) is a method by which biomechanical rupture risk can be estimated based on patient characteristics and a computed tomography (CT)-derived 3D model of an AAA. Microarray analysis allows high-throughput analyses of tissue gene expression.

Aims: The overall aim of this thesis was to explore and develop new strategies to improve, refine and individualize management of patients with AAA, by applying geometric, biomechanical and molecular analyses.

Methods and Results: In study I, the CTs of 146 patients with AAAs of diameters between 40 and 60 mm were analyzed with three-dimensional (3D) segmentation and FEA. Simple and multiple regression analyses were performed. Female sex, patient height, lumen volume, body surface area (BSA) and low BMI were shown to be associated with the biomechanical rupture risk of AAA. Study II included 191 patients with AAAs of diameters between 40-50 mm. The AAAs were analyzed with 3D segmentation and FEA after which prediction algorithms were developed by use of machine learning strategies. More precise diameter measurements improved prediction of growth and four-year prognosis of small AAAs. Biomechanical indices and lumen diameter were predictive of future rupture or symptomatic AAA. Growth and rupture required different prediction models. In study III, 37 patients, 42 controls and a validation cohort of 51 patients were analyzed with respect to their circulating levels of neutrophil elastase-derived fibrin degradation products (E-XDP). The results showed that E-XDP was a sensitive marker for AAA, independently of examined comorbidities, and its concentration in peripheral blood correlated with the AAA diameter and the volume and mechanical stress of the intraluminal thrombus (ILT). It was further increased by the presence of coexisting aneurysms. Study IV included 246 tissue samples, divided into tunica media and adventitia, from 76 patients with AAA and 13 organ donor controls, analyzed by microarrays. There were large differences between the transcriptomes of AAA and control media and adventitia. Processes related to inflammation were transmural, whereas the upregulation of proteolysis, angiogenesis and apoptosis along with downregulation of smooth muscle- and differentiation-related gene sets were specific for the aneurysm media. Active smoking increased oxidative stress in all tissues and increased inflammation and lipid-related processes in AAA. The growth rate of the AAA diameter correlated with adaptive immunity in media and lipid processes in adventitia.

Conclusions: In this thesis, we show that known clinical risk factors and certain geometric properties are associated with biomechanical deterioration of AAAs. Furthermore, geometric and biomechanical analyses can enhance prediction of outcome. Importantly, there are differences between prediction of AAA growth and rupture. Finally, a biomarker was discovered and the transcriptome of AAA including effects of the ILT, smoking and rapid diameter growth rate, was mapped and we envision that the data may be used for future biomarker and drug target discovery.