The impact of aging and atrial fibrillation on thrombus formation in-silico

Abstract Background Atrial fibrillation (AF) is an age-related cardiac arrythmia that underlies one third of all strokes, with thromboemboli commonly originating from the left atrial appendage (LAA). Personalised in-silico modelling of Virchow’s triad (blood stasis, hypercoagulability and endothelial damage) can improve stroke risk assessment metrics, such as the CHA2DS2-VASc score, by providing quantitative mechanistic information [1]. The ability to quantify the impact of aging may play a critical role in patient stratification and therapy selection to reduce the risk of AF-related stroke. Aim We aim to perform in-silico modelling to quantify the effect of aging and AF on thrombus formation in the LAA by using advanced computational tools to capture blood flow, clotting mechanisms and endothelial damage [2]. Methods A virtual physiological LA model was derived from cardiac Cine MRI from an adult AF patient imaged during sinus rhythm (SR). Biomarkers for a geriatric patient were obtained from literature and applied as: i) elevated plasma fibrinogen concentration and ii) longer thrombin time-to-peak (ttP) [3]. Blood flow was simulated in SR and AF conditions for 5 cardiac cycles in the LA for both ages (totalling 4 test cases). The catalytic conversion of fibrinogen to a fibrin clot by thrombin was modelled to mimic hypercoagulability [2]. The site of endothelial injury was prescribed in the middle of the LAA. The area under curve for the fibrin concentration (Fn AUC) was measured to quantify thrombus formation. Results In the two SR cases, LA blood flow velocity increased to a maximum of 1.8 m/s after 0.6 s with a mean of 0.63 m/s during the cardiac cycle, while the two AF cases showed 1.3 m/s and 0.39 m/s, respectively, corresponding to a 38% decrease in AF. Peak velocity in the LAA was 0.35 m/s during SR and 0.18 m/s in AF. The fibrin concentration in the adult-SR case rose to 1.1 mmol/m³ over the first cardiac cycle, but both thrombin and fibrin were washed out of the LA after 1.5 s, with a Fn AUC of 0.045mmol s/m³ (Fig. 1a). The geriatric-SR case sustained fibrin generation throughout the 5 cardiac cycles reaching 3.1 mmol/m³, an Fn AUC of 2.4 mmol s/m³, and the formation of an LAA thrombus (Fig. 1b). Similarly, the adult-AF and geriatric-AF cases also saw continuous growth to 5.0 and 7.4 mmol/m³, with Fn AUCs of 6.4 and 12.6 mmol s/m³, respectively, with formation of a thrombus in the LAA in both (Fig. 1c and d). The thrombus formed in the adult-AF case was 25% larger than the geriatric-SR case, while thrombus in the geriatric-AF case was 80% larger. Conclusions This investigation quantifies the increased risk of LAA thrombus formation in older patients, with an Fn AUC increase of 55% between adult and geriatric in SR and a 100% increase when combining geriatric biomarkers with AF conditions. With further validation on larger patient cohorts, this novel in-silico tool may improve our understanding of the risks in AF-related stroke.