AF is the most common heart rhythm disorder, responsible for approximately one-third of hospitalizations for cardiac rhythm disturbances in the US. The prevalence and incidence of AF are increasing. It is predicted to affect 6-12 million people in the US by 2050 and 17.9 million in Europe by 2060, significantly impacting wellbeing and healthcare costs. AF is associated with increased morbidity and mortality, due to the risk of ischemic stroke, systemic embolism, heart failure, and cognitive impairment, reducing the quality and quantity of life in these patients (4). This condition is associated with a six-fold increase in stroke. Moreover, patients with previous ischemic stroke are at an even higher risk (5). In the case of cardioembolic strokes (two-thirds of the total), echocardiographic and pathologic studies suggest that when a source can be identified, approximately 90% of such strokes can be attributed to thrombus formation in the left atrial appendage (2). Non-rheumatic AF is the most frequent source of cardioembolic brain infarct (57.1% of cases) followed by valvular heart disease (20.3%) and coronary artery disease (18.2%), that AF can also occur in patients with atherothrombotic stroke as epiphenomenon or as clinical manifestation of atherosclerotic disease (16.5%). In-hospital mortality in patients with atrial fibrillation was significantly higher than in non-atrial fibrillation patients both in cardioembolic (32.6% vs. 14.8%, P<0.005) and atherothrombotic stroke (29.3% vs. 18.8%, P<0.04). Valvular heart disease, and sudden onset (OR 1.8; 95% CI 0.97–3.63) were predictors of cardioembolic stroke, and subacute onset, (Chronic obstructive pulmonary disease), hypertension, hypercholesterolemia, transient ischaemic attack, ischaemic heart disease, diabetes of atherothrombotic stroke (6). Bayés syndrome (Bayés de Luna 1988) is a new clinical entity, characterized by the association of advanced interatrial block (IAB) on surface electrocardiogram with AF and other atrial arrhythmias. This syndrome is associated with an increased risk of stroke, dementia, and mortality. Recent studies have shown that Bayés syndrome is a key independent factor of cardioembolic cerebral ischemia[, although there is still a need for a high level of clinical suspicion in order to diagnose it. Early and proper diagnosis of Bayés syndrome is desirable and necessary since patients will require closer clinical surveillance, and possibly accompanying antiarrhythmic and antithrombotic preventive therapies. The clinical relevance of Bayés syndrome lies in the fact that is a clear arrhythmological syndrome and has a strong association with supraventricular arrhythmias, particularly atypical atrial flutter and atrial fibrillation. Likewise, Bayés syndrome has been recently identified as a novel risk factor for non-lacunar cardioembolic ischemic stroke and vascular dementia (7).

The primary approach to prevent ischaemic stroke risk associated with AF, are anticoagulation therapies, which have a proven efficacy. At least four large clinical trials have clearly demonstrated that anticoagulation with warfarin decreases the risk of stroke by 50–80% (8–11). In relatively recent trials, the newer oral anticoagulants (OACs), such as dabigatran, rivaroxaban, apixaban, and edoxaban, have proven to be similarly effective to warfarin for the prevention of stroke and thromboembolism (12). However, anticoagulants have significant drawbacks. Although anticoagulants reduce 30-day mortality from ischaemic stroke, these agents increase intracranial haemorrhage-related mortality (13). Moreover, patients having a stroke despite being on therapy with an oral anticoagulant are at high risk of recurrent ischaemic strokes (14). In the ANAFIE registry, patients at high bleeding risk had higher incidences of stroke, major bleeding, intracranial haemorrhage, gastrointestinal bleeding, cardiovascular events, and all-cause death than patients in the reference group, despite a high prevalence of OAC therapy (89.0%) (15). In elderly patients, non-adherence to OAC treatment, associated commorbidities and additional risk factors can significantly increase the incidence and severity of cerebrovascular accidents. In this age group, a delicate balance may exist between multiple conditions, being thrombotic disease, chronic kidney disease, cancer, coronary artery disease, and heart failure, some of the most challenging scenarios encountered in clinical practice (15). In addition, warfarin OACs have multiple contraindications (16): ∙

prior intracranial haemorrhage or diseases predisposing to intracranial haemorrhage,

TAVR has emerged as an alternative, rapidly evolving non-invasive procedure for patients with severe aortic stenosis and medium-to-high surgical risk. By 2025, there will be an estimated 280,000 TAVR procedures performed worldwide and the total market will exceed $8 billion. Although this highly promising treatment modality results in less morbidity, shorter time to recovery and similar mortality rates, it is still associated with one of the most devastating and feared complications: cerebral embolism, which in turn may cause stroke (17). Stroke is associated with a 6-fold increase in mortality in TAVR cohorts, a moderate to severe permanent disability in up to 40% of survivors (5), a 4.7-fold increased risk of permanent work disability (2), social isolation and significant financial strain in 80% of stroke survivors (13), and an increased risk of readmission in patients with stroke after cardiac catheterization (18).

The time in between the TAVR procedure and the cardioembolic event is an important factor when choosing stroke prevention treatments. Most of them occur in the acute phase following TAVR where cerebral embolic events are frequent (19). Nonetheless, according to the ADVANCE trial, half of the reported strokes occurred between day 2 and 30 after the TAVR procedure (20). Moreover, evidence is mounting on ischaemic brain lesions being produced after day 30, causing SBIs and long-term neurological symptoms. New ischaemic brain lesions were found in 74% to 100% of patients on diffusion-weighted magnetic resonance imaging (DW-MRI) after TAVR (21). Although studies have shown that SBI may not be related to apparent short-term neurological symptoms, evidence points to an association with accelerated cognitive decline and strengthening of the risk of long-term dementia (most commonly, Alzheimer’s disease) (22). Later events are associated with patient specific factors (23). SBIs have been associated with accelerated cognitive decline and higher risk of long-term dementia. The situation is especially worrying given that TAVRs are being implanted in increasingly younger and lower risk patients, hence potentially increasing the prevalence of dementia.

As mentioned, in patients undergoing TAVR, stroke remains a potentially devastating complication associated with significant morbidity and mortality. This is especially worrying, given that TAVR is now indicated in aortic stenosis low- and intermediate-risk patients (24), and that stroke rate 30 days post-TAVR has been reported at 3.4% in low risk-patients (25). To prevent debris from embolizing to the brain during the procedure and reduce the risk of stroke, cerebroembolic protection devices (CEPDs) were developed (26). These devices are implanted during the procedure and up to 2 days after, since the clogging of CEPDs impedes them from being used to prevent ischaemic strokes >2 days post-TAVR. Nonetheless, as explained above, risk of stroke may not be limited to the procedure itself or the perioperative period. Moreover, the clinical benefit of current CEPDs in reducing strokes, transient ischaemic attacks or death remains unknown (27–29). It is therefore worrying that for strokes triggered by AF or other conditions extended in time, there are no currently approved CEPD in the market. The SENTINEL-LIR study demonstrated that embolic debris captured by the SENTINEL-CPS (Cerebral Protection System) during TAVR in low- to intermediate-risk patients was similar to that in previous studies conducted among higher-risk patients. These findings suggest lower-risk patients undergoing TAVR have potentially a similar embolic risk as high-risk patients, as evidenced by embolic debris capture (30). Most captured debris had a size of <500 μm, with 78% between 150 and 500 μm. Larger size particles (≥1000 μm), which can cause significant vessel obstruction, were present in 67% of cases. Therefore, the Sentinel CEPD can functionally capture large debris that may cause a severe stroke. In contrast, debris on the micrometer scale may pass through the gaps between the filters and arteries, leading to stroke even in CEPD-protected territories (with less likelihood of severe symptoms) (31). Dedicated meta-analyses demonstrated that SBIs of small infarction volume 3 mm3 are independent predictors of later stroke and mortality (18), further highlighting the need for a CEPD able to filter small debris in the long term post-TAVR. Such a device would be a radical improvement for a large population at risk.

The main purpose of the work presented in this paper can be described with the simple motto: Save the brain. The proposed solution consists of a stent filter/diverter attached to an electronegatively charged strut structure, intended to block the passage of clots or deflect their trajectories thanks to both fluid- and electro-dynamical forces on the clots. This can be achieved thanks to the electrostatic repulsion acting on naturally electronegatively charged blood clots, and its geometrical design, capable of filtering thrombi based on the distance between struts. The main objective of this device is to reduce or eliminate the percentage of blood clots entering the BCT, LCCA, and LSA, while maintaining at least 98% of the natural delivery of oxygenated blood to the brain. The Section 4.3 will demonstrate that the presence of the devices in all three arteries reduces the flowrate in a maximum of 7.5% in the case of BCT, and in less than 2% in the other two aortic arch arteries. This paper provides the details of the work carried out using Computational Fluid Dynamics (CFD) coupled with Lagrangian particle transport to validate the efficacy of the deflecting medical device positioned at the base of the BCT, LCCA, and LSA. A sufficiently large number of particles were injected into the domain in order to assure significant statistical samples for validating the particle-deflecting capabilities of the device. The study is divided in two main stages. The first one analyzes the effect of the thickness and shape of the strut design on the device performance. In the first stage, the purely hydrodynamic effect of the device is analyzed using a CFD and a particle transport model. The device is placed at the root of the LCCA and the optimal strut thickness is identified by analyzing the trajectories of particles suspended in the flow. The analysis showed a low efficacy for the deflection of thrombi and identified a deficiency in the initial design which is was not avoiding particles pass through the lateral struts. To overcome this deficiency, extra struts are added in the second design employed in the second stage of the work, oriented perpendicularly to the original struts. The second stage of this project, proposes to simulate struts that are electrically charged on their surface. Considering that blood clots are negatively charged, the strut surface would be negatively charged too in order to repel the clots. The proposed device consists of a stent NITINOL, plus Graphene oxide/ Bovine serum albumin nanoparticles (Bovine serum albumin bioconjugated graphene oxide: Red blood cell adhesion and hemolysis studied by QCM-D .Bing Cai) attached to an electronegatively charged strut, preventing the adhesion of cells and blood proteins, as well as reducing the possibility of hemolysis to minimal. The design is intended to block the passage of clots or deflect their trajectories being possible to deflect even the smallest particles more than 50% from entering the aortic arch arteries, offering a method for decrease Silent Brain Stroke. This can be achieved thanks to the electrostatic repulsion acting on naturally electronegatively charged blood clots, and its geometrical design, capable of filtering thrombi based on the distance between struts. The main objective of this device is to reduce or eliminate the percentage of blood clots entering the BCT, LCCA, and LSA, while maintaining at least 80% of the natural delivery of oxygenated blood to the brain.

The project has not yet been tested on animals or patients; these phases are planned for the future. Our initial prototype was developed to demonstrate proof of concept. The most recent design, which is nearing completion, is self-expanding and can be placed inside a 10-French catheter for delivery. In terms of deployment, CEPDs are implanted either during the procedure or up to two days afterward. After this period, the potential clogging of the CEPDs makes them unsuitable for preventing ischaemic strokes (>2 days post-TAVR). The risk of stroke may extend beyond just the procedure or perioperative period. One of the significant advantages of our Sentinel CEPD is its capability to capture large debris that could cause severe strokes. However, smaller debris on the micrometer scale might still bypass the filters, leading to potential strokes in CEPD-protected areas, albeit with a reduced chance of severe symptoms. Meta-analyses have indicated that small-volume SBIs (<3 mm3) can predict subsequent strokes and increased mortality, underscoring the long-term need post-TAVR for a CEPD designed to filter out even small debris. Such a device would mark a significant advancement for a large at-risk population.

The effectiveness of CEPDs in minimizing strokes, transient ischemic attacks, or mortality remains to be determined (18). It’s concerning when considering strokes induced by AF or other extended conditions.

For device residency duration, we should cater to two distinct populations: the elderly group with AF, or those immediately post-stroke to avoid anticoagulation. For both groups, the device would be permanently placed. For TAVI patients classified as low, medium, or high risk without AF, the device should remain for longer than 30 days, acting as a CPD, and based on their treating physician’s discretion. However, for TAVI patients who either had AF before the procedure or developed it post-procedure, it’s recommended to keep the device in place permanently.

The aortic arch geometry used for computational simulations was designed based on an anatomical model from a previous study (32). In order to adapt the geometry to the requirements of the present study, some modifications were made using ANSA v22.1.0 (Beta Simulation Solutions, Lucerne, Switzerland). The applied modifications preserve the original structure of the aortic arch geometry, only in the LCCA and BCT significant changes (i.e., go from an anatomical structure to a synthetic one) have been applied. It was observed that the original geometry was not representative of an average BCT geometry, so it was manually corrected to adapt it to a more realistic one. Likewise, it was noticed that the LCCA presented a broader morphology at the entrance of the vessel, which could imply obtaining wrong conclusions, thus, it had to be corrected. Finally, let us comment that the aortic root was truncated to focus on the vessels in the aortic arch. For the inlet, the area corresponding to the left ventricle outflow tract (LVOT), with the coronary arteries and the aortic valve was replaced by a rigid tube to ensure the correct stabilization of the fluid. At the outlet, the area corresponding to the superior mesenteric, illiac, and renal arteries was dismissed. The resulting geometry used in this study can be observed in Figure 1.

For the second stage of the present work, the device is implanted at the base of the three arteries: BCT, LCCA, and LSA. Figure 2 shows the patient-specific geometry of the aortic arch employed, with the arteries where the device will be placed. The inlet and outlet boundaries are labelled as: ∙

Inlet: Aortic root

The subsequent results comes from the two stages of the presented work: ∙

For the first stage, with no electrical forces and as it has been described in Section 4.3 and illustrated in Figure 4B, the analysed particle deflecting efficacy is focused on the behaviour of the geometric design parameters: the distance between deflector struts or strut interval (Lsi) and the strut thickness (Lst). The deflector is only located at one artery, LCCA, and 5 different prototypes based on mentioned parameters are analysed and presented in the Supplementary Appendix.

In this section the device performance is analyzed for the healthy flow rate conditions, in the baseline (i.e., no deflector), deploying a single deflector on each one of the aortic arch arteries separately, and finally deploying the deflectors in all arteries at once.

In this section the device efficacy is analyzed for the AF flow rate conditions, in the baseline (no deflector, Figure 8A), deploying a single deflector on each one of the aortic arch arteries separately (Figures 8B–D), and finally deploying the deflectors in all arteries at once (Figure 8E).

One of the main necessary conditions to assure the safety of the device, is that the presence of the deflectors does not reduce the blood flow rate significantly into the aortic arch arteries, necessary to provide the brain with oxygenated blood. Figure 11 overlap the flowrate curves at these arteries, in the baseline case (without deflector) and in the case where the three devices are deployed simultaneously, both for the healthy and AF patients. These correspond to the worst case scenarios for reduction of the flow, and therefore are sufficient to assure the device safety when the device is deployed in a single or two arteries. From these curves, it is observed that the device does not alter the flow rates significantly in any case. The maximum reduction is found in the BCT artery, with a 8.63% reduction for AF patients and 8.64% for healthy ones, with respect to the baseline case. In the LCCA artery the reduction is equal to 1.61% and 1.59% for AF and healthy patients while for LSA there is 4.03% and 3.97% less flow for the AF and healthy patients respectively when the device is deployed. In summary, the model predicts that the device reduces the ejected flowrate to the aortic arch arteries by significantly less than 20%, which is the maximum tolerated physiological reduction. Therefore, the device complies with the necessary condition set at the beginning of this report, which is required to assure a healthy cerebrovascular function of the treated patient.

Particles blocked by the mere fact that their diameter is larger than the inter-strut spacing may cause concern regarding clogging of the device, as occurs with conventional CEPD. The current numerical model cannot evaluate this aspect since deformability of the particles is not modelled. Nonetheless, it is somewhat reassuring that in contrast to conventional CEPD, the current device has a positive convexity which slightly bulges into the main aortic domain. This geometric attribute combined with the high velocities in the aorta result lower the possibility of particles getting stuck in between struts.

This effect is also addressed and improved introducing the repulsive electrical force due to the inverse distance proportional factor presented in this force.