Abdominal aortic aneurysm (AAA) is a ‘silent killer’, responsible for 15,000 deaths annually in the UK alone. This cardiovascular disease is termed a ‘silent killer’ due to its deadly but somewhat symptomless nature. The majority of people with AAA are completely unaware of its presence and it is often only detected following a frequently fatal rupture of the structure. AAA is a balloon-like dilation occurring in the abdominal region of the aorta, which is the main blood vessel responsible for carrying blood from the heart to the rest of the body (Fig 1). If undetected or untreated, AAA can continue to expand over time and can eventually rupture. Rupture of AAA is a catastrophic event causing major internal bleeding and is associated with a high mortality rate. At present, the risk of death due to rupture is of great concern: 90% of patients die before reaching the hospital and the survival rate for those who do make it to the hospital is only 35-70%. [login type="readmore"] As a result, AAA is currently the 13th leading cause of death in the United States and has claimed the lives of many famous people in the past, including Albert Einstein and former French president, Charles De Gaulle. Interestingly, AAA is four times more likely to occur in men than in women, with men over the age of 65 years being the most high-risk group. Conversely, the risk of rupture is about four times higher in women. [caption id="attachment_2798" align="aligncenter" width="612"] Fig 1. Location and formation of abdominal aortic aneurysm[/caption]
The exact reasons for AAA development are not yet fully understood. However, it is thought to be linked to the primary factors influencing cardiovascular disease in general – for example, tobacco smoking, sex, age and high blood-pressure. Due to the lack of symptoms associated with AAA, detection is difficult and often only occurs by chance when investigating an unrelated condition.
Upon detection of AAA, there are two primary methods of surgical repair: the traditional open repair and the minimally-invasive endovascular repair (EVAR), which both involve placement of a prosthesis (graft or stent-graft) inside the AAA, relieving the pressure on the weakened AAA wall and returning the blood flow to a relatively normal state.
The timing of surgical intervention is non-trivial and the assessment of rupture risk is now high on the clinical research agenda. When AAA is detected, surgeons have to assess whether the risk of rupture is greater than the risk of surgery (surgical mortality rate: 1-6%). Currently, the risk is considered high enough to warrant surgery when the AAA is over 5.5cm for men (a slightly lower threshold is usually considered for women) and has grown 0.5cm in a year. However, research shows that this assessment is unreliable and does not account for reported rupture in smaller AAAs and non-ruptured large AAAs. It is known that about 20% of AAA < 5cm rupture and that as many as 75% of AAAs may never burst. Therefore, many patients have unnecessarily faced the risks of surgery while others have been offered treatment too late. It is generally understood that is a need to develop a tool to aid surgical assessment of AAA rupture risk. RUPTURE RISK ASSESSMENT Currently, rupture risk assessment is a ‘one size fits all’ process and fails to consider that no two AAAs are the same. From a basic mechanical viewpoint, the wall of the AAA ruptures when the blood pressure is greater than the strength of the AAA wall. Similarly with a balloon, if the pressure inside exceeds the strength of the balloon wall it will burst. At the Centre of Applied Biomedical Engineering Research (CABER) at the University of Limerick, we believe that by assessing the in vivo wall stress of a patient’s AAA (i.e. patient-specific) and comparing this to the strength of AAA wall, we can help surgeons isolate patients who have a high rupture-risk, regardless of AAA size. The wall stress of a patient-specific AAA can be calculated by reconstructing a 3D virtual model of the patient’s AAA from computed tomography (CT) scans using specialised software. Then using the advanced engineering tools of finite element analysis (FEA), we can predict the wall stress acting on the AAA (Fig 2). For this processing chain, three pieces of information are required: (i) Patient-specific geometry Upon detection of AAA, a CT scan is routinely performed to plan surgery. Using image processing software called Mimics (Materialise, Belgium), we can create a patient-specific 3D virtual model of the AAA geometry from these CT images (Fig 2 (a)). This model also includes the intraluminal thrombus (ILT); a complex multi-layered structure attached to the inner AAA wall in 75% of all cases (Fig 3(a)). The 3D model is then converted into a finite element mesh for analysis. [caption id="attachment_2802" align="aligncenter" width="1630"] Fig 2. Finite element method: (a) complete reconstruction of AAA geometry from CT images; (b) each 3D reconstruction consists of outer AAA wall and ILT (if present); (c) material properties are assigned, boundary conditions are applied and FEA is performed; (d) resulting stress contours on AAA wall[/caption] (ii) Material properties A critical aspect of the development of these models is defining the material properties of both the AAA tissues (Fig 2 (b)), that is, the ILT and the diseased AAA wall. In order to model them accurately, we need to determine how tissue behaves in response to pulsatile blood pressure. At CABER, we are currently utilising/developing two methods: 1. In vitro: biomechanical tests, which aim to mimic the in vivo loading, are performed using tissue removed from patients undergoing open repair surgery at the HSE Midwestern Regional Hospital, Limerick. An equivalent pressure to that within the body is applied to the tissue, and stiffness and strength are measured (Fig 3). [caption id="attachment_2809" align="aligncenter" width="1375"] Fig 3. Determining material properties in vitro (a) human ILT harvested from AAA sac; (b) 9mm square samples prepared for testing; (c) biaxial testing of ILT; (d) Resulting stress stretch relationship which is used to determine material properties such as strength and stiffness[/caption]
2. In vivo: the pulsatile AAA can be imaged in vivo using state-of-the-art techniques such as acoustic radiation force imaging (ARFI) and cardiac gated CT. By examining the movement of the AAA wall during a cardiac cycle, it is possible to calculate the material properties.
(iii) Boundary conditions and applied loads AAAs are typically located between the renal arteries that pass blood to the kidneys and the iliac arteries that run down each leg. These arteries prohibit the AAA from deforming at the proximal and distal ends. Similar boundary conditions are applied to our computational models. To simulate the internal pressure of the blood acting on the AAA, either the patient’s exact blood pressure can be measured at the time of CT and applied to the inner wall of the model, or to compare multiple AAAs, a standardised pressure of 120 mmHg is used. Each computational model is assessed for mesh independence to ensure accurate results without unnecessary computational time. STRESSES ON AAA WALL AND WITHIN ILT The result is a detailed description of the various stresses acting on both the AAA wall and within the ILT (Fig 2 (d)). The major findings of our group, some of which have been published in scientific journals, include: • Virtual models can estimate with reasonable accuracy in vivo wall stress and by comparing this to wall strength can identify high rupture risk AAAs; • Material properties can influence wall stress predictions, hence there is a need for improved characterisation of this diseased tissue; • The ILT is a patient specific, inhomogeneous, isotropic material and should be modelled accordingly; • The accuracy of wall stress estimates can be improved by the inclusion of ILT in the model; • Certain shaped AAAs produce higher wall stress than others; • Development of rupture prediction tools such as finite element analysis rupture index (FEARI), which couple in vivo wall stress to in vivo wall strength, can help surgeons isolate high-risk AAAs. Based on the work by both our group and others, we believe a tool capable of identifying high risk AAA will: (i) Increase standard of treatment Some 10-24% of ruptured AAAs are smaller than 5cm; despite this, current risk assessment criteria exclude them from treatment due to their size. The proposed tool will identify high-risk AAA regardless of size, treatment will be received on time, potentially saving the lives of many people. (ii) Reduce unnecessary surgeries Currently, hospitals are performing 10 operations to prevent one rupture. It is possible, by isolating high risk AAA, nine out of 10 operations may be avoided. The benefits of this include: • Reduced surgical risks including heart attack, stroke, kidney failure, infection and spinal cord injury; • Reduced post-surgery risks including graft infection, leakages and failure; • Reduced hospital costs: in Galway University Hospital alone, this would have resulted in a saving of over €2 million over the last five years. POTENTIAL BENEFITS The potential benefits of the development of this tool are worldwide, potentially saving the lives of thousands of people whilst dramatically reducing unnecessary surgical risk and associated costs (billions/year worldwide). Detection of AAA is improving; as a result, numbers of reported AAAs are increasing. It was recently estimated that AAA diagnosis has tripled over the last 30 years and currently over 500,000 new cases are diagnosed worldwide each year. In light of this, we realise now more than ever, the importance of this research and its potential to dramatically improve our standard of care for those unfortunate enough to develop this deadly disease. Siobhan O’Leary, Centre of Applied Biomedical and Engineering Research (CABER), Department of Mechanical, Aeronautical and Biomedical Engineering, Materials and Surface Science Institute (MSSI), University of Limerick, Ireland In support of this research, the author would like to acknowledge (i) the Irish Research Council for Science, Engineering and Technology for the provision of funding, (ii) the Higher Education Authority for providing the necessary equipment, (iii) clinical collaborators at the HSE Midwestern Regional Hospital, Limerick: Mr Eamon Kavanagh, Prof Pierce Grace, Prof Stewart Walsh and Mr Paul Burke and (iv) supervisors Dr Barry Doyle and Prof Tim McGloughlin for their excellent support and guidance.