Development of a novel computational framework to support therapeutic-planning in selecting the optimal thromboembolic prevention treatment for AF

点击翻译按钮获取中文摘要

1) Brief description of the context of the research including potential impactAtrial fibrillation (AF), the most prevalent cardiac arrhythmia, is characterised by rapid and disorganised heartbeats. It affects 2-3% of the population, with prevalence rising to 9% for individuals > 65 years. It is the leading cause of thromboembolic events (i.e. stroke and vascular dementia), and its projected incidence is more than double by 2035.It is estimated that > 90% of thrombi responsible for thromboembolic events under AF originate in the left atrial appendage (LAA), a complex-shaped protrusion of 2-4 cm, which departs from the left atrium (LA).The most common therapy to prevent thrombi in AF patients is oral anticoagulation; this is however contraindicated in up to 44% of AF patients, due to associated lifetime haemorrhagic risk, cost and several food/drug interactions needing frequent laboratory monitoring. Alternatively, surgical LAA exclusion can be performed applying an external suture ligation - e.g. LARIAT snare device (SentreHEART) - or a self-closing clamping device - e.g. AtriClip LAA Exclusion System (AtriCure). Percutaneous LAA occlusion devices overcome the drawbacks of invasive surgery, based on self-expanding nitinol frames supporting polyester occlusion patches. The two most common solutions are the Watchman FLX (Boston Scientific) and the Amplatzer Amulet (Abbott). However, these can develop peri-device leaks, requiring long-term anticoagulants treatment, thus inherently defeating the main procedure purpose.As all these options present major drawbacks, it is crucial to develop a novel approach to identify the patient groups for which the risk of thromboembolic events justifies a specific therapeutic approach, and the safest and most effective therapy for each subject.2) Aims and ObjectivesThe overall project goal is developing a novel computational framework to support clinical decision-making in selecting the optimal thromboembolic prevention therapy for each AF patient, identifying fluid-dynamic parameters that are impossible to measure in-vivo. 1. Analyse and segment in-vivo medical images of AF patients to obtain LAA's 3D morphological/dynamic characteristics.2. Develop Fluids Structure Interaction (FSI) computational models for the AF population, including LAA/LA anatomical features, investigating the relations between different LAA morphologies and hemodynamics, as well as the capturing of the ones not available from clinical diagnostics.3. Develop models of the main thromboembolic prevention treatments (i.e. anticoagulants and LAA excluder and occluding devices), allowing the evaluation of anatomy-treatment interaction and fluid-dynamic conditions.4. Validate the computational models with retrospective clinical data.3) Novelty of Research MethodologyDevelopment of a validated FSI computational framework, here specifically applied to AF patients, to select the most suitable therapeutic strategy to prevent thromboembolic events among those currently available: anticoagulation drugs, surgical/percutaneous occlusion of the LAA.4) Alignment to EPSRC's strategies and research areasThe project aims at developing a novel computational framework to support therapeutic-planning in the selection of the optimal thromboembolic prevention treatment for AF patients. This will be based on fluid dynamic parameters impossible to measure in-vivo, and will require the synergy between recent developments in medical imaging and computational engineering. It aligns ideally with the EPSRC's strategic priorities of patient-specific illness prediction, accurate diagnosis, and also transforming health and healthcare.5) Any companies or collaborators involvedProfessor Pier Lambiase, Co-Director of Cardiovascular Research Barts NHS Trust/UCL Institute of Cardiovascular Science (ICS)Dr Oliver Segal from Cardiac Units at UCLH and the Barts Health NHS TrustDr Claire Martin