Abstract Number ‐ 244: Feasibility of Detecting Carotid Stenosis Microemboli Using Autonomous Robotic TCD at a Large Stroke Center

Abstract

Introduction There is accumulating evidence that the presence of microembolic signals(MES) could be an independent risk factor and marker for future stroke or TIA. These signals detected by ultrasound represent small particles (platelets,plaque,fibrinogen) too small to cause a neurologic deficit, but can be a warning of a larger particle dislodging and blocking an intracranial artery. Transcranial doppler (TCD) studies are historically small. Few technicians have the specialized training to perform them and experienced users often have difficulty obtaining an adequate echogenic window. This study utilizes an autonomous robotic transcranial doppler (aTCD), driven by artificial intelligence to obtain and optimize the cerebral blood flow velocity signal. This tool makes data acquisition more efficient thus increasing analytical power through ease in obtaining larger sample sizes. This study focuses on the process of integrating such a tool. Complex vascular cases can utilize such a tool for patients on the borderline of needing a carotid stent or endarterectomy to help ensure the selection of optimal surgical candidates through better risk assessment methods. Methods NovaGuide™ 2 Intelligent Ultrasound (NovaSignal Corp. USA) was utilized to scan patients for microembolic signals that met our criteria. Our population consisted of male and female inpatients that were at least 18 years old at a single center with a CTA head and neck that revealed 50% or greater internal carotid artery stenosis for patients with neurological symptoms that could be associated with this artery. For patients that were asymptomatic with an incidental finding of internal carotid artery stenosis, we included 60% or greater stenosis. Stenosis was measured by NASCET criteria. If a patient received a stent or carotid endarterectomy during their hospitalization, a microembolic scan was obtained before and after their procedure if the initial scan was positive. Scans were performed for at least 30 minutes if the patient could tolerate this. Results Data collection is ongoing with a plan to scan 20 patients from July 2022 to October 2022, but the initial 12 patients revealed no MES which is contrary to the rate found in other studies. Potential confounders include antiplatelet and anticoagulation use as these were not delayed if clinically warranted. All cases were symptomatic and carotid stenosis ranged from 50% to 99%. A request for echo technician feedback is pending, but initial comments have revealed 30 minutes to be the maximum time feasible at our institution. Conclusions The aTCD has shown in a prospective study (NCT04604015) that data can be collected by non experts. Our study demonstrates that such a tool can be integrated into a large stroke center by echo technicians not formally trained in manual TCD. Utilization of technologies that allow for simplified data collection can help power larger TCD studies to explore its use as a risk utilization tool. This can help with selection of ideal surgical candidates for carotid stent and endarterectomy as an additional data point in their overall case. The significance of MES requires further investigation, but studies utilizing conventional TCD support MES being a risk factor for stroke and TIA.